Madagascar - Forests Management and Protection Project

Document of The World Bank FOR OFFICIAL USE ONLY Ch'74_ MAcrC Report No. 6994-flAG STAFF APPRAISAL REPORT DEMOCRATIC REPUBLIC OF MADAGACAR FORESTS MANAGEMENT AND PROTECTION PROJECT January 14, 1988 Agricultural Division South Central and Indian Ocean Department, Africa Region This doeument has a restricted distribution and may be used by recipients only In the perfonnance of their officbl duties Its contents may not othewise be disclosed without Worid Bank authorization.  CURRENCY EQUIVALENTS Currency unit Malagasy Franc (FMG) US$1.00 FMG 1,360 FMG 1,000 = US$ .74 (as of July 1987) WEIGHT AND MEASURES metric British/US Equivalents 1 meter (m) - 3.3 feet (ft) 1 kilometer (km) .62 mile (mi) 1 square kilometer (kM2) = .39 square mile (sq.m.) 1 hectare (ha) 2.47 acres 1 kilogram (kg) G 2.20 pounds (lb) 1 metric ton (-.ton) 2,205 pounds (lb) GOVERNMENT ADMINISTRATION Fokonolona Institutions Fokotany village Firaisam-pokotany - group of Fokotany (or Firaisana) (former canton) Fivondronam-pokonola group of Firaisana (or Fivondronana) (former sub-prefecture) Faritany - Group of Fivondronana (former Province) FISCAL YEAR Governmmnt of Madagascar January 1 - December 31 Fanalamanga July 1 - June 30  VOR OFFCIL USE ONLY DEMOCRATIC REPUBLIC OF NADAGA8CAR Forests Nanatement and Protection Prolect Glossary of Acronyms and Abbreviations AIVF AmSnagement Int;gro des Valloes Forestiores Integrated Forest Valley Developmont ASAC Credit d'Ajustement du Sectour Agricole Agricultural Sector Adjustment Credit ABC Contre Appui Agroforestior Agroforestry Support Center AVF Amenageaent des Valloes Foresti;res Forest Valley Development AWSA Bass in d'Approvisionnement Bois d'Antananarivo Antanaarivo Wood Supply Area IADKA Banque Arabs pour le D;voloppoment Economique de l'Afrique Arat Bank for African Economic Development BII Bankin' My Indostria National Bank for Industry CFPF Centre de Formation Professionnelle Foresti4re Forestry Professional Training Center CU Uniti de Coordination Coordination Unit DAWS Division Appui Suivi at Statistiques Monitoring and Statistics Division DEF Direction des Eaux et Forats Forestry Department DEPF Direction des Etudeo, Programes at Financ_ment (du MPAEF) Department of Stud4e-s, Programs and Financing (of MPAEF) DEL Direction de l'Elovage Livestock Department DIR Direction de l'Infrastructure Rurale Rural Infrastructure Department DR"P Direction de la Recherche Foresti4re et Piacicole Forestry and Fisheries Research Department DVA Direction de la Vulgarisatiou Agricole Agricultural Extonsion Department KASTA Ecole d'Application des Sciences et Techniques Agricoles Agricultural Technical School SESSA Etablissement d'Enseignement Sup4rieur des Scienees Agronomiques Agricultural Sciences University FIUUULUSPNGA Fanjarian' Ala Ambatondrazaka-Moramanga Mangoro Forest Company (Projoct Entity) FFN Fonds Forestior National National Forestry Fund FHP Plan d'Aminagement Forestior Forest Management Plan FIPA Zone de Planification do l'monagomont Foreatier Forest Management Planning Area FNDE Fonds National D voloppem_nt at Equipement National Investment Budget FOFIFA Foibem Pironona Momba ny Fykarohana Ampibarina smintny This decumet has a restitd distdbuton and may b usd by reipit, ony in u pforuanue of ti offiW duies. its contents may not otrws be discsd withu Wodd ank autoidati.  Fampandrosoana ny Ambanivohitra Agricultural Research Center GoM Gouvernement de la R"publiquo Democratiqu. de Madagascar Governmsnt of the Democratic Republic of Madagacar IUCN International Union for the Conservation of Nature Union Internationale pour la Conservation de la Nature ICB International Competitive Bidding LCB Local Competitive Bidding MPAEF Ministere de la Production Animale et des Eaux et Forets Ministry of Livestock, Fisheries and Forestry MPARA Ministere do la Production Agricole et do la Reforme Agraire Ministry of Agricultural Production and Agrarian Reform MUE Suivi et Evaluation Monitoring and Evaluation HMDC Cooperation Norvegienne Norwegian Ministry of Development Cooperation PC Coordinateur du Projet Project Coordinator PEP Programm do Dpensoes Publiques Public Expenditures Program PIP Programme d'Investissement Public Public Investment Pro;rem PPF Avance de Fonds de Preparation Project Preparation Facility SDC Cooperation Suisse Swiss Development Cooperation SIEF Service de l'Inventaire et de l'Exploitation Foresti;re Inventory and Forest Exploitation Service SPEF Service Provincial des Eaux et Forkts Regional Forestry Service SPN Service de la Protection de la Nature Nature Protection Service TSC Centre Appui Technique Technical Support Center UNDP Programme des Nations Unies pour le D;veloppement United Nations Development Progr&. WwF Fonds Mondial pour la Nature World Wildlife Fund ZODAFARB Zone Delimite d'Action en Favour do l'Arbre Reserved Private Reforestation Area  DEDOCRATIC REPUBLIC OF MADAGASCAR Forests Hanaatment and Protection Proloot Credit and Project Summary Borrowers Democratic Republic of Madagascar Beneficiarys Ministry of Livestock, Fisheries and Forestry Amounts SDR 5.1 million (US$ 7.0 million equivalent) Terms: Standard 1/ Cofinancierst Cofinacing would be provided from a grant from the Swiss development Cooperation (SDC) of US$ 6.5 million equivalent and a grant from the Norway Ministry of Development Cooperation (NMDC) of US$ 2.0 million eq4uivalent. IDA would be the administrator of both grAnts. Prolect Descriutiont Project objectives are to help the preservation of ecosystems and biologicAl species in Madagascar and assist In Implementing the National Conservation Strategy and Forastry Policy, adopted respectively in 1984 and 1985. To this ond the project seeks to: (a) strengthen the Forestry Department so as to manage the sector and priority program; (b) prevent further degradation of natural forestsg and (c) promote private sector Involvement in reforestation and wood processing. The project would include the following three components: (a) Porestry Department reinforcement through institutional strongthening and human resource development activities, and support to the forest management, private reforestation, integrated forest valley development, seed collection and research programs; (b) natural forest protection of the Ankarafantsika, Lokobe, Zahamena and Tsaratanana forest reserves, and creation of a new National Park in the Andasibo region; (c) reorientation of the Mangoro plantation (Fanalemanga) including sylvicultural program for long term sawlog production. The project would be carried over a period of 7 years. Benefits and Risks: Ecological benefits would arise from the preservation of ecosystems and biological species unique to Madagascar. 1/ On-lending terms between the Government of Iadagascnt (GOH) and Fanalamanga for US$2.4 million will be with a maturity of 20 years, Including a grace period of 7 years and an Interest rate 10 higher than the IBRD rate at the time of Board presentation, the foreign exchange risk being born by Fanalamanga.  - II - Effective forest management and agroforostry programs would provide long term Impacts on agricultural and forest productivity including conservation of productive lands. Benefits would arise from the development of the Forestry Department human resources, planning and sector management capacity. Incremental production from project activities would include fuelwood, sawlogs, fooderops and specific tree products such as fruits and resin. Distributional benefits to poor rural households would arise from the integrated forest valley development and natural forest protection programs. Main project risks include insufficient enforcement related to forest management activities leading to depressed wood-product prices, weak field execution capacity and inadequate management of sector and project activities at senior levels. Project emphasis on annually agreed work programs, on task and geographical selectivity, and on human resourco development aim at reducing these risks.  - iii - Bstlisted Project Coat Local Foreoln Total ^--------US$ million------- DEY Reinforcement 2.3 5.5 7.8 Natural Forest Protection 1.4 4.7 6.0 Reorientation of Mangoro Plantation L-1 2.5 6.4 Total Baseline Coats 7.6 12.7 20.3 Physical Contingenties .2 .5 .7 Price ContSngenciea .6 1.0 1.6 Total Project Costs 11 8.4 14.2 22.6 Financing Plan IDA .7 6.3 7.0 SWISS COOPERATION 1.2 5.3 6.5 NORWAY COOPERATION 2.0 2.0 Go0 and Cost Recoveryl 6.5 7.1 TOTAL 8.4 14.2 22.6 11 Includes taxes and duties of US$ 2.89 million and PPF of US$.9S million 21 Includes taxes and duties of. US$ 2.89 million (:32 of total cost). est5Aated Disburstmonts (US $ million) IDA FY 199 1990 1991 1992 1993 1994 1995 Annual 2.0 1.4 .6 .7 .7 1.5 .1 Cumulative 2.0 3.4 4.0 4.7 5.4 6.9 7.0 Economic Rate of Return of incremental investments in the MNagoro Plantation (Fanalamanga) tabove 50%. Staff Appraisal Re&ort Report No. no. 6994 NAG MaD IBRD 20657  - iv- DEMODCRAIC REPUBLIC OF MADAGASCAR Forests Manatement and Protection Proiect Staff A.nraisal Report Table of Contents Pate No. Credit and Project Sumary ............. ..... ..... .9..... i-iil Table of Contents..................... .................... iv-vi I. BACKGROUND A. Introductton .................... 1 B. Economic and Agricultural Context ................ 2 C. Governmont Agricultural Policy Fr*Aework .... .... 3 D. Bank Group Assistance to the Agricultural Sector... 4 Z. Bank Role in the Forestry Sector................. 6 II. FORESTRY SECTOR A* Forest 7....................................99 7 B. Environmental Aspects... .................... 8 C. Marketis and Reforestation ........................ 8 D. Institutional Setting ....................... ..... 10 F. * vermnt. Policy e ................ ................. 11 P. Sector Firtance ............ 11l III. THE PROJECT A. Project Strategy and Objectives.................. 13 B. Project Design and Summary Description**..*.*..., 13 C. Geographical Coverage ............................ 14 D. Detailed Project Features........................ 14 Forestry Department Reinforcement ............. 14 Natural Forest Protection ..................... 20 Reorientation of the Mangoro Plantation ...... 21 Technical Assistance and Training ...999.... 22 IV. PROJECT COST, FINANCING, PROCUREMENT AND DISBURSEMENT A. Project Costs .......................... 99 9999999 22 B. Project Financing ................................ 23 C* Procuremen5t.............. . 25 D. Disbursements .................................... 26 E. Accounts and Audits.............................. 27  o,v - V. PROJECT MANAGEMENT AND InPLEMENTATION A. Project ........... ***... .**. * ... 28 8. Implementation of Project Componentso......... 28 DIF Reinfore.ment......... ................... 28 Natural Forest Protection.......ecton....... 32 Reorientation of Mangoro Plantation ...... 33 C. Annual Work Program...... 34 D. Monitoring and Evaluationo..o .............os* 35 VI. PRODUCTION, MARKETS, PRICES AND FINANCIAL ASPECTS A. aroetso. ...... ..* 36 Bo akt 37 Co Pc .e e s 39 D. Financial Analysis of Fanalanan. s 40 Z. Cost Recovery and Fiscal Impact.................. 41 VII. BENFITS, JUSTIFICATION AND RISKS A. Bo nfenefits... ...... .........aa ...............a.. 42 B. Environmental Imaa e t 43 C. Economic Analysis (Mangoro Plantation)..o........ 43 D. Sensitivity Analysis (Mangoro Plantation) ....... 44 E. Project Risks............... . 45 VIII. AGREDEENTS TO BE RACHED A C H E D o*o,*** ******* .... 46 TABLES IN MUIN TEXT Table 2.1 Forest Ares" by Faritany................. 7 Table 4.1 - Proj*ct Costs Summary.................... 23 Table 4.2 - Financing Plsn 24 Table 4.3 - Procurement Procedures. dures............. 25  -vi ANEXN S Annex 1 DEF Staffing 1987 51 Annex 2 .Table 1 - Operating Budget 52 .Table 2 - INDE Investment Budget 53 .Table 3 - Public Investment Program 1987-89 54 Annex 3 .Current Protected Areas Activities 55 Annex 4 .T.chnical Assistance Requirements 56 Annex 5 .Terms of Reference for major Technical Assistance 57 Annex 6 .Project Related Training 70 Annex 7 .Table 1 - Project Components by Year 71 .Table 2 - Summary Accounts Cost Sumar 72 .Table 3 - Breakdown of Sumuary Accounts 73 .Table 4 - Summary Accounts by Year 74 .Table 5 - Sunnary by disbursement category 75 Annex 8 .Table 1 - Procurement Proceduress DEF Reinforcement 76 .Table 2 - Procurement Procedures: Natural Forest Protection 77 .Table 3 - Procurement Procedures: Reorientation of the Mangoro Plantation 78 Annex 9 .Table 1 - Proceeds SDC grant IDA Credit: DEF Reinforcement 78 .Table 2 - Proceeds IDA Credit SDC and NMDC grants: Natural forest Protection 80 .Table 3 - Proceeds IDA Credits Reorientatica of the Mangoro Plantation 88 Annex 10 .Schedule of Disbursements 82 Annex 11 *Implementation Schedule 83 Annex 12 .Staffing Plan 34 Annex 13 .Field Activity Monitoring Assisaments 85 Annex 14 .Financial Analysis of Fanalamanga 86 Annex 15 .Economic Analysis of Fanalamanga 48 Annex 16 .Cost Recovery 100 Annex 17 *Government Cash Plow 104 Annex 18 .Working Papers 105 CHARTS 1. Current DEF Organizational Chart 106 2. Proposed DEF Organizational Chart 107 3. Fanalamanga Organizational Chart 108 4. Coordinating Unit 109 5. MPAM? Organizational Chart 110 MAP IBRD Activity Location Map (# 20657) 111  DEMOCRATIC REPUBLIC OF MADAGASCAR FORESTS MANAGEMENT AND PROTECTION PROJECT I. BACKGROUND A. Introduction 1.01 Forests in Madagascar are of particular environmental and economic significance due to their high level of biological diversity and to their watershed protection role in a country with a high susceptibility to erosion. The uniqueness of these forests combined with the high rate of deforestation have brought the attention of the Government of Madagascar (GOM) and the international community to the need for urgent action to improve the management of wood resources and forest protection. GOM has consequently formulated and adopted a National Conservation Strategy in 1984 and a Forestry Policy in 1985. Within the basic thrust of these policy documents, the proposed project provides specific support to a mtually agreed priority action program. 1.02 Project Background. The proposed project is the outcome of an ongoing dialogue between 1OM and the Bank on forestry sector strategy. Initial project preparation reflected an awireness of growing urban needs and a dwindling accessible fuelwood energy supply. A March 1984 FAO/CP preparation report consequently focussed on fuelwood production. Based on the conclusions of a subsequent joint Bank-FAOICP preparation mission in May 1985, which benefitted from the preliminary findings of the Energy Assessment mission, agreements were reached with GOM on modifications in project scope and objectives. These agreements reflected a conviction by both GOM and the Bank that a forestry sector approach would be more appropriate than the originally restricted fuelwood plantation focus. An institutional development component was prepared with IDA assistance, and a FAOICP mission returned to Madagascar t! May 1986 to consolidate the findings of several short term consultancies for the Mangoro plantation and to develop a rural forestry component. A pre-appraisal mission was present in the field in February 1987. The International Union for the Conservation of Nature and Natural Resources (IUCN) and the World Wildlife Fund International participated in the mission to assist in the preparation of the natural forest protection component. The project was appraised in June 1987.1 B. Economic and Agricultural Context 1.03 Economic Context. With a population of ten milliotn inhabitants and a per capita income of US$ 265 in 1986, Madagascar Is one of the I/ The appraisal mission consisted of J. Tanaka, M. Grut and B. Ribon (IDA). Ms. M. Harris, ME. U. Park and Mr. T.N. Rico-Mora (consultant) assisted with the preparation of the report.  poorest countries in the world. The izland has an area of 590,000 km2, a significant and varied natural resource base and a population density of about 17 inhabitants/km2. Inappropriate economic policies, inefficient administrative controls and overinvestment based on inadequate borrowing arrangements in the late 1970's led to a difficult period of financial stabilization in the 1980's. Economic growth in the first half of the 1980's was insufficient to keep pace with population increase. Measured in constant prices, the estimated per capita GDP in 1985 was only about 80% of the level achieved in 1980. Overall deficit of the central government has, however, been reduced from 18% of GDP in 1980 to an estimated 5% in 1985, and the balance of payments current account from nearly 20% of GDP in 1980 to about 10% in 1985. In parallel to its commitment to continued stabilization measures, Government's new policy has been direated towards increasing exports, encouraging economic efficiency, enhancing competition, reducing administrative controls, and adopting measures that would allow market forces to play an increasing role in resource allocation. The recently (June 1987) approved Industry and Trade Policy Adjustment Credit supports these policy reforms by inducing major trade liberalization and exchange rate adjustment. 1.04 Agriculture Sector Outline. Agriculture is Madagascar's dominant economic sector; it employs 85Z of the population and accounts for more than 80S of export earnings. Production and farming systems vary widely, with many commodities produced, but production in value terms is dominated by paddy (30%) and beef (16%). A small group of export crops (coffee, cloves, and vanilla) accounts for about 15% of production, but some three- quarters of merchandise exports. Industrial crops, notably sugar and cotton, contribute about 7% of agricultural production value. Approximately 80% of total production comes from smallholder activity, and small farmers (1 to 2 ha range) are the backbone of Madagascar's agricultural economy. Over half of agricultural production in value is for subsistence, and the share of subsistence production has slowly increased over the past decade. In contrast to most sub-Saharan African countries, irrigation is widely practiced in Madagascar, with water zontrol systems in use on about 1 million hectares (one third of cultivated land). Overwhelmingly, irrigated land is used to produce rice, Madagascar's main crop. The country's rich and varied agricultural potential offers excellent prospects for raising incomes, improving the supply of industrial inputs, and for increasing foreign exchange earnings, but this potential is far from being realized. 1.05 Past Performance. Agricultural sector performance has been poor since the mid 1970s. In the 1975-78 and 1984-86 periods, production growth was insufficient to keep up with population growth and between 1979 and 1983, agricultural production stagnated. Most serious problems have been: (a) the stagnation of rice production and the very serious decline in edible oils production; and (b) the decline in coffee exports (40% of merchandise exports) by about a third from peak quantities which has not been compensated by increased export of other commodities. One of the few brighter spots has been increased cotton production between 1983 and 1986 but the serious price decli oe in world markets has recently led to major pricing and marketing problems which may be now eased after the June 1987 devaluation (para 1.03).  1.06 Constraints to Growth. The disappointing performance of the agricultural sector since 1972 resulted from a package of government interventions which effectively disengaged the sector from the rest of the economy, and pushed a large part of it back into subsistence. Government suppressed or eliminated private trade and as the sole buyer for key crops offered very low prices to producers. Physical barriers to trade, erected mostly by local and provincial governments, curtailed the emergence of a parallel market which in other African countries has rescued the sector from the worst effects of policy intervention. The growing misalignment of the currency intensified price distortions, especially for export crops. Tariff protection and the growing inefficiencies of manufacturing worsened the terms of trade for agriculture as a whole. With the assistance of the Bank this situation is now being unwound. Since 1984 markets have been progressively liberalized, and in the case of export crops the enormous gap between producer and world prices has been partially closed. More income is flowing to rural producers, but the sector is not yet restored to its proper funct'oning and place in the economy, and, except pe.haps for rice, there is as yet no evidence of a production response. Part of the ongoing difficulty reflects the evolution of the import regime which is not yet free; the continuing shortage of foreign exchange despite recent devaluations, and the fact that industrial/manufacturing tariff protection is still high. Partly it reflects the fact that the trade sector is not yet competitive or efficient. Nominal free entry Is slow to take effect because of the suppression of the financial system, and very large profit opportunities and mark ups ar. needed to elicit even a small trader response. The imminent reforms of the Import regime, the tariff regime, and the domestic financial system, along with a restructuring of agricultural and export taxes, represent the next stages of the recovery process; these issues are being tackled under the current program of policy loans. C. Governmenc Agricultural Policy Framework 1.07 Strategy. The agricultural sector crisis, especially the decline in marketed output of food and industrial commodities and shrinking exnort levels, led to a fundamental rethinking of Government strategy in the early 1980s. A new policy framework was formally presented at the April 1983 Consultative Group (CG) Meeting and subsequently confirmed at tae April 1986 CG meeting. The Government's objectives now are to (a) increase production of food crops (above all, rice and edible oil crops) in order to reduce imports; (b) revive production and improve the quality of traditional export crops; and (c) diversify agricultural exports. Better integration of the agricultural and industrial sectors has become an important medium-term objective. The strategy to achieve these objectives also recognizea that small and larger farmers are the main agricultural producers and that Government policy and investments must create an environment which stimulates production by these key economic agents. Equally important, the new strategy recognizes that the private sector has a positive role to play in the process of economic and agricultural development. The practical consequences of this approach are that the GOM now supports: (a) moving towards a free market system as a means of improving producer incentives; and (b) withdrawing ministry services from  activities better performed by other operators, whether small farms (e.g. agricultural seed production) or commercial companies (e.g. input supply). Public investments in the sector are to be limited to rehabilitation over the medium term, and systematic efforts are to be made to program and manage them better. 1.08 Stratety Implementation. Over the past three years, the Government with Bank Group support has made steady progress in implementing this new strategy. Most significant actions have involved the removal of agricultu,al pricing and marketing controls on rice, meat, and pulses, and increased farmgate prices for cotton and selected export crops. In relation to resource management, GOM has prepared and reviewed with IDA a Public Investment Program (P.I.P) for agriculture, azid is establishing systems to update the program annually. Parastatal operations are progressively bel'-! subjected to profitability tests. GOM is also reinforcing extension and research while withdrawing from direct production and some commercial activities. The agricultural ministries, the Ministry of Agricultural Production and Agrarian Reform (MPARA) and the Ministry of Livestock, Fisheries and Forestry (MPAEF) have been substantially reorganized, and management strengthening programs launched. Personnel reductions have been undertaken to position them to play a more dynamic but less interventionist role. 1.09 Sector Adjustment. As a counterpart to the reduction in Govornment intervention and the reduced role of public enterprises in the economy, Government has taken actions to promote development of the private sector in order to reduce the bias which emerged during the years of economic dirigism. The steps taken so far include (a) the adoption of a new Investment Code in mid-1985 which improves legal and fiscal incentives to private businesses; (b) easier access to credit and foreign exchange; and (c) the elimination of state marketing montopolies for major crops, including paddylrice, beef and beans. Further measures envisaged includet (a) a program to foster the development of commercial marketing circuits; (b) the elimination of price discrimination against the private sector in the marketing and processing of export crops; and (c) the handover of agricultural input and veterinary products import and distribution to private firms. Taken together, these measures add up to the first phase of a solid adjustment package which will hopefully revive business confidence and promote private sector investment in the development of agriculture. D. Bank Group Assistance to the Agricultural Sector 1.10 Relation to Government Stratety. Bank Group support for agricultural sector development in Madagascar has been comprehensive, including project financing, sector work, institutional development support, and an intensive dialogue with the Government on key policy issues. The Bank Group has also played an important role in helping the Government to coordinate the activities of external financing agencies in agriculture by encouraging co-financing arrangements, by its review of the Public Investment Program (PIP) (para 1.08) and through the vehicle of Consultative Group (CC) meetings (para. 1.07).  1.11 Bank Suvoort to Agriculture. As of July 1987, agricultural lending to Madagascar accounts for about US $234 million for 16 operations over a 14-year period, out of a total Bank Group portfolio of about US$700 million. Eight Bank financed agricultural projects are currently under implementation, in addition to an IFAD project appraised and supervised by Bank staff. Bank Group lending has been primarily concentrated on priority commodities, such as rice, livestock, and cotton and has sought to address the critical sector issues outlined above (para. 1.06). Since 1983, lending has become more directly program and policy oriented. 1.12 Experience with Bank Lending. Eight completed projects have been audited by the Operations Evaluation Department. Project performance has been mixed, with most projects affected both by policy and institutional shortcomings concerning the whole sector, and by the repercussions of the recent economic crisis. The main lessons which have emerged from the lending experience are that (a) new technology needs to be carefully adapted to local circumstances, if necessary by trial and error, which calls for considerable design flexibility in agricultural projects; (b) administrative capaci.y is very weak and even with support can be easily overwhelmed by attempting too much too soon; (c) parastatals are a very poor substitute for private traders in providing marketing for farmers and by their inherent weakness they can worsen the price and income which risks farmers face; and (d) without properly functioning markets, competitive private commerce, and reasonable terms of trade for the sector, the chances of successfully introducing new technology and encouraging increased production are not high. 1.13 Recent Operations. The Bank Group has responded to specific project issues, as wvll as to the sectoral/economy-wide problems outlined in para. 1.06 througt& intensive supervision of on-going operations, sector work, and preparation and appraisal of a new generation of projects. The lead operation in addressing sector policy issues is the first Agriculture Sector Adjustment Credit approved in May 1986. This operation provides a credit of SDR 16.5 million for needed imports of agricultural inputs, veterinary products and tractors and transport equipment, and incentive goods for rural areas. The adjustment program to which the credit is tied couprises the following main elements: (a) activities to improve market efficiency through continued liberalization and development of the private sector; (b) strengthening producer incentives by relying increasingly on market forces, while improving the Government's remaining intervention in pricing; (c) setting up an effective rice management program which will include ending the Government's marketing in the two major rice surplus zones; (d) rationalizing public expenditure, notably through a better agricultural investment plan; (e) irrigation rehabilitation project -- complementary investments to the rice program; and (f) preparing the next phase of the agricultural reform program. A key component of this operation is the rice security strategy designed to provide incentives for increased domestic production and including a rice intervention stock to stabilize prices during the scarcity period. Its implementation so far has led to an important increase in domestic rice production, making Madagascar self-sufficient in rice in 1987. In addition, recent devaluations have made Malagasy rice sufficiently competitive to enable a resumption of better quality rice exports to the regional market. Another important measure is the increase of coffee farmgate prices so as to expand coffee production, Madagascar's principal export.  1.14 Institutional Develoument and Credit. The Second Agricultural Institutions Development Project, also approved in Msy 1986, is designed to improve the efficiency of public sector institutions and follows from the successful first phase project which provided institution building support notably to MPARA and MPAEF. Specific objectives are to improve resource management, strengthon policy analysis capability, and develop management capacity for key Governmont services responsible for extension, research, data collection and analysis. In addition, specific studies will contribute to policy analysis and preparation of the next phase of the reform program. The Second Agricultural Credit (Cr 1804-MAG) project is designed to address the third key element in redressing Madagascar's agricultural production, namely to help private entrepreneurs invest in the agriculture sector. It also provides funds to support further institutional strengthening messures for the National Rural Development Bank (BTK) and implement initiatives to reform the banking sector under the Industry and Trado Policy Adjustment Credit. Z. Bank Role in the Forestr"' Sector 1.15 HanM oro Forestry Prosect (Loan 1065-MAGa Cr. 525-HAG). Under the First Mangoro Forestry Project (1974-81), 47,000 ha of industrial pine plantations for the production of woodpulp were established by Fanalamanga, the parastatal created to implement this project. Though the major physical objective of the project was exceeded by 35Z (12,000 ha planted over appraisal estimates), it became apparent shortly after project completion that, for various reasons including deficiency of zinc (Zn) as a trace element and potassium. (K) as a major element, a high proportion of the plantations (mostly those created prior to the project period) was not growing as expected, and that yields from these plantations would be far below expectations. At the same time the wood pulp market slumped and has not yet fully recovered. The Project Performance Audit Report (no. 4198, dated November 14, 1982) analyzed main project issues, and noted some important lessons, most significant of which was the need for effective monitoring mechanism. 1.16 Second Manaoro Forostrv Prolect (Cr. 1161-MAG). This 'ollow-up project was initially designed to further sustain plantation ac;-vities. Because of doubts about tho feasibility and justification for the pulpmill, the project was redesigned at appraisal to reduce planting rates, to cover only a three-year period and, most important, to incorporate studies of options for industrial utilization of plantation products. A GOM technical review committe concluded in March 1984 that the best option was a phased, domestic market-oriented, integrated development program with emphasis on saw-milling and small plants for production of kraft pulp, particle board, and plywood and to reduce nw plantations. Fanalamanga has been to date an effective project implementation agency with strong management and a qualified work force. It has vigorously pursued the agreed fertilization and planting measures, has started a thinning and pruning program to ensure that the final harvest will be of sawlog quality, and has implemented improved fire protection and control measures. Due to the above mentioned  - 7 - reorientation after the 1984 review and the subsequent severe reduction of its planting program, it Is currently estimated that by the extended closing date of December 31, 1987, US$ 3.7 million will remain undisbursed. 1.17 With the two above Forestry Projects, the Bank role in the Forestry Sector was focussed on industrial forestry parastatal exploitation. It has become apparent that the Bank should reorient its assistance to the forestry sector and have a more global approach in terms of institution building and prcgrasming, forest exploitation control, assistance to villagers for reforestation, expansion and protection of natural reserves, and limit the role of the Mangoro plantation to the production of standing trees for sale to private enterprises. Because of tho broader view encompassed in this project and the reorientation of the Mangoro plantation, the project name has been changed from Third Forestry Project to Forests Management and Protection Project.2 II FORESTRY SECTOR A. Forest Resources 2.01 An estimated 12 million ha or one fifth of Madagascar's territory remain covered by natural forests. Only 141 of theso forests are considered to be high density non-degraded forosts. Wet evergreen forests are found along the eastern escarpment, while western forests are essentially deciduous or semi-dociduous. These forest ecosystems rely on a fragile reproduction equllibrium, and, while it is believed that the island was once mostly covered by forests, it is today undergoing rapid deforestation. It is estimated that the forest cover is decreasing by 200,000 ha a year due primarily to clearing for agriculture and forest exploitation. There are currently an estimated 260,000 ha of plantation forests, predominantly consisting of pine and eucalyptus. Estimates of natural forest and plantation areas by faritany are presented in Table 2.1. Table 2.1: Forest Areas by Faritany (ha) Faritany Natural Plantations Total Forests Antananarivo 114,500 60,929 175,429 Antairanana 1,504,300 5,510 1,509,810 Fianarantsoa 1,285,000 77,609 1,362,609 Mahajanga 2,147t,400 6,691 2,134,091 Toamasina 2,813,700 102,106 2,915,806 Toliara 4,462,000 11,932 4,473,932 Total 12,306,900 264,777 12,571,677 2/The previous title Third Forestry Project may still appear on several Tableaux.  B. Environmental Aspects 2.02 Environmental Significance. Madagascar is endowed with an evolut.ionarily and biologically unique environment, having been isolated from other land masses for some 60 million years. It has the highest rate of biological endemism in the world: all the lemurs, 95X of the reptiles, 86% of the plants and over half of the bird species are found only within the island. Madagascar has thus become a top priority area for the international conservation community due to the rate of ecological degradation and the particular significance of preservation of biological diversity. 2.03 Soil Erosion. While Madagascar has a moderate average real rate of erosion of 1.38 mm/year, it has a high rate of potential erosion at 4.05 mm/year indicating that the real rate may increase substantially in the absence of preventive measures. This is particularly noticeable for the eastern zone of Madagacascar where the current real average rate of erosion is estimated at .71 mm/year while the potential rate reaches 6.04 mm/yr. Increases in the real erosion rate lead to a deterioration of soil fertility which threatens the long term agricultural potential of Madagascar's land resources. It also causes siltation in irrigation networks, induces the further clearing of new lands, and favors landslides and flooding during the rainy seasons which destroy roads, valley crops and buildings. Deforestation is an important cause of the erosion problem. It has a grave impact on a large portion of the remaining natural forests located in the major eastern and northwestezn river watersheds as rorest degradation leads, for example, to loss of topsoil and severe siltation in the two major rice producing areas of the Lac Alaotra and Marovoay. Reforestation, forest protection and soil conservation programs must form part of a strategy to address this issue. C. Markets and Reforestation 2.04 General Wood Market Characteristics. Annual wood consumption is estimated at some 10 million tons, 9 million tons of which is consumed as a source of household energy. Fuelwood accounts for 96% of national household energy consumption and for some 802 of total final energy consumption. The geographical distribution of forests is inversely proportional to that of the population. The Antananarivo region contains 30% of the population and less than 1% of the remaining natural forest area. According to analyses performed by the Bank's Energy and Industry Operations, the supply-demand imbalance for fuelwood in the Antananarivo region will increase by 70% over the next decade from the current deficit of 1.3 million tons of wood equivalent. This would correspond to a loss of 1.5 million ha of natural forest cover by 1995. Paradoxically, fuelwood prices over the last five years appear to have decreased in real terms, reflecting a breakdown in the Forestry Department capacity to control forest exploitation as well as a significant drop in consumer purchasing power and increased competition in fuelwood supply. The valorization of  -9- existing wood resources appears in this context as an essential condition for the rehabilitation of the forestry sector (para. 3.11). Annual Lndustrial roundwood consumption is estimated at 1 million tons with 552 of the market concentrated in Antananarivo. The sawlog market has more than doubled between 1968 and 1986 in spite of the economic slump. This is attributed to price increases of competing construction materials (such as corrugated iron sheets and cement) and the growth of the Antananarivo market. The share of pine wood has also more than doubled in that period reaching an estimated 401, as wood from the natural forest becomes less accessible (para. 6.05). While the first two forestry projects misjudged the market capacity to absorb Mangoro plantation output (woodpulp), the proposed project is demand-driven. However, since the growth of the domestic sawlog market is dependent on the uncertain growth of the economy, market forecasts, for the Mangoro plantatior. are conservative both in volume and in prices (para 6.04-6.09). 2.05 Export. Export revenues from the forestry sector have decreased by 75X in constant terms over the last ten years. This decline has been mostly caused by a reduction of the external market for natural fibers (such as raffia or kapok). Exports of other products have either stagnated or decreased with the exception of medicinal plants and cinnamon. Between 1972 and 1974, foreign exchange earnings from exports balanced foreign exchange expenditures for imports of forestry related products (mostly paper and pulp). The trade balance specific to the forestry sector has been consistently negative since then. This deficit reached 2.8 billion FMG (US$ 4.5 million) in 1983. Madagascar could develop its export market for products such as medicinal plants, cinnamon and sawnwood (para. 6.06). 2.06 Industrial Plantations. Madagascar, with the support of the Bank, UNDP, BADEA and other donors, launched an ambitious plantation program in the Mangoro valley designed to support future major forest-product industries. These industries would have included primarily a pulpmill, and also sawnwood and charcoal production. Due to the slump in the world pulp market, the orientation of this plantation has had to be shifted from an export pulp production focus to a longer-term sawlog production scheme. 2.07 Pzivate Reforestation. Farmers have played a minor role in wood and tree related production. Community-based Government sponsored programs have failed until recently because they relied mostly on coercion for inducing tree planting and because lack of enforcement of forest exploitation regulations has led to depressed wood market prices. Rural forestry represents an untapped potential in most regions of the country for both tree-related subsistence products and for domestic and export markets products. In order to better utilize this potential, GOM has introduced, in decree 85-072, a new approach to private reforestation denominated ZODAFARB (Zones delimitees d'action en faveur de l'arbre). This approach is based on three principles: first, that tree planting should be a profitable investment; second, that the trees should belong to the planter; and third, that title to designated areas of state land should be given to individuals or groups who plant and maintain trees on them. Response to the ZODAPARB program, especially from businesses, associations and individuals from Antananarivo has so far been encouraging with survival rates of planted trees reaching 851 after the first year.  - 10 - D. Institutional Settint 2.08 Institutional Structure. The Forestry Department (DEF) of the IIPAEF plays a central role in the conception and implementation of the national forestry policy. The DEF current main functional responsibilities include forest management (exploitation control and inventory), nature conservation (protection of forest reserves and soil conservation) and reforestation (see Chart 1, actual organization chart of DEF). Regional Forestry Services constitute the decentralized implementation structure of the DEF in the field. A network of forest protection areas has been established and consists of 2 national parks, 11 strict natural reserves (with total protection) and 23 special reserves covering at its creation an area of 1.2 million ha. Natural forests not included in the network of protected areas are mostly part of the public domain. Their exploitation, as a first priority, Is regulated by the DEF which is responsible for granting exploitation permits, determining stumpage rates, and on-site exploitation control. The DEF staff comprised 868 persons in 1987 of which 62 were at the engineer level (professional level with undergraduate degree in forestry) and 42Z at the technical level. Annex 1 presents the distribution of DEF staff by grade and faritany. 2.09 While the DEF's organizational structure is basically correct, and its legal framework to act is adequate, it has in practice been very weak in fulfilling its role. First, its budgetary resources have been substantially reduced in real terms, thus compressing both the investment budget and the operating budget severely (see para. 2.13). Second, staff have not been renewed for a long time, as a result of which some 40X of agents are over 50 years of age. Moreover, many positions go unfilled due to the closing of technical schools between 1981 and 1986. For instance, only 372 of field network positions are currently filled. This has led to DEF's incapability to supervise correctly the forest protection areas, which has resulted in continued encroachment, and has also made control of forest exploitation virtually non-existent. The proposed project will attempt a comprehensive reinforcement of the DEF by (a) implementation of a staffing plan for project activities; (b) strengthening its programming, program monitoring and financial management; (c) implementation of a broad training program for local staff; a-id (d) selective technical assistance to reinforce the DEF is priority tasks. 2.10 The two major forestry parastatals are Fanalamanga (Mangoro pine plantation) and Famama (cashew plantations and processing) with several other smaller parastatals involved in wood-processing activities. Most wood processing parastatals are currently operating significantly under capacity. The private sector deals in all aspects of forestry production from plantation to wood processing. Charcoal production is a profitable activity for small scale operators, but would be hardly profitable for large operators. There are few private large scale wood processing enterprises as most sawmilling is still done by family owned small scale businesses.  - 11 - E. Government Policy 2.11 Forestry Policy and Objectives. A Forestry Policy and Development Strategy has been formally adopted by GOM in 1985. Main objectives include: (a) increase forested areas through participatory private reforestation programs; (b) rationalize the management of existing forests; (c) reduce land area affected by fire and cleared for agriculture; (d) define a watershed management plan; (e) step up enforcement of protected natural areas; and (f) reach fuelvood self-sufficiency for major urban areas. The formulation of these objectives represents an important step in guiding the development of the forestry sector, although the level of operational specification remains insufficient. The policy does not provide an indication of the priorities to be pursued in the short term. This represents a serious shortcoming given the current limited means of the DEF. 2.12 Conservation Policy. The Government has recently started to address conservai.ion issues with the promulgation of a Conservation Strategy decree, the creation of a National Conservation Commission chaired by the Prime Minister, and the sponsoring of a major international conference in November 1985 to increase the awareness of conservation issues in Madagascar and attract financial support from donors. The implementation of this strategy has been s"w and the resources of the Government are clearly very limited to address this problem. The proposed project would contribute to the implementation of several Important elements of the strategy, such as: promoting reforestation by inducing private participation, village-based agroforestry programs, intensifying cultivation In suitable areas (irrigated rice in forest valleys), and promoting tourism based on respect for the country's unique natural heritage. F. Sector Finauce 2.13 Sector Finance and PIP. An examination of operating and investment budget allocations reveals a dramatic drop in the financing available to the sector. In constant terms, for the foreatry sector, the operating budget has decreased by 692 since 1977 (Annex 2, Table 1), and the investment budget by 752 since 1980 (Annex 2, Table 2). Personal expenditures represent 932 of the operating budget and over 601 of the investment budget. An examination of the 1987-89 Public Investment Program (PIP; Annex 2, Table 3) for the forestry sector raises the following issues: (a) the level of programmed resources for the sector is unrealistic in relation to past financing levels. The financing gap is considerable as only 9% of requested external grants have been acquired. Requests for national investment budget allocations for the 1987-89 period are equivalent to an average annual FNDE allocation of FMG 4 billion whereas the 1987 allocation was equal to FMG 1.44 billion;  - 12 - (b)in spite of the relative growth of other programs since the 198 4-87 PIP, the largest share of planned resources is still too concentrated on industrial plantation projects (53Z of the PIP). The Mangoro pine and cashew plantations alone represented 80% of external financing committed to the sector in 1984. Industrial plantation projects received 95% of committed external financing and 701 of the national investment budget. This contributes to increase the financing gap for other sector activities. This project devotes 301 of total project costs to the Mangoro plantation which will become profitable in year 7 and hence will allow GOM resources to be directed into other sectors. (c)the PIP consists of 28 projects ranging from FMG 52 million to FMG 4.39 billion. The eight smallest projects represent only 5% of the PIP. The PIP would benefit from the consolidation of activities into a coherent framework of clear investment objectives leading to the identification of clear investment priorities. Certain activities, which have proven insufficiently productive after several years of investment should be investigated and, if need be, reoriented. The shift in emphasis of the proposed project from an industrial plantation focus to a broader sectoral approach structured around a limited number of specific priorities would help PIP regain balance by responding appropriately to key forestry sector issues and by attracting financing. In addition, the development of a Public Expenditures Program (PEP) within the context of the proposed project would help identify additional issues intrinsic to the DEF operational programs/project activities relationship. The project will assist the DEF to revise its PIP and PEP. 2.14 National Forestry Fund. The concept of a National Forestry Fund (FFN) has been introduced by decree in 1985 to strengthen sector finance. Funds for the FFN would initially consist of an annual FNDE allocation. Such a measure would allow to have a budgetary cycle appropriately timed with reforestation activities since FNDE funds are currently available only from May to December (due to a lengthy budget implementation procedure) whereas the peak of the planting season runs from January to March. GOM envisages that FFN sources would eventually include all revenues from forestry exploitation permits and taxes on wood product transactions. The FFN would focus initially on the ZODAFARB program (para. 2.07) and would progressively support all DEF activities as sources of funds are expanded. Because the FFN has not yet been established, the project has been designed in a way that does not require it. III. THE PROJECT A. Proiect Strategy and Obiectives 3.01 Strategy. The proposed project represents a reorientation from previous Bank-financed fcrestry activities in Madagascar (para 1.,7). First, it draws away from a narrow industrial plantation focus and addresses broader forestry sector issues with an emphasis on conservation and sustainable use of natural resources. Second, it assists in redefining  - 13 - GOP's role in the sector away from direct wood production and processing towards more effective provision of incentives for reforestation and improved forest management. Third, it supports an increased role of the private sector in wood production by assisting participatory private reforestation programs. This approach is consistent with the agricultural strategy outlined in para. 1.07 and responds to the Bank Group commitment to assist Sub-Saharan Africa in dealing with deforestation, fuelwood supply and conservation issues. It is also consistent with the policy of emphasizing the private sector since Fanalam'nga would limit its role to selling to private entrepreneurs either fell wood from pruning (for charcoal), or standing trees for sawing. 3.02 Oblectives. Project objectives are to: (a) strengthen the DEF capacity to manage the sector and execute priority programs to implement the strategy (para. 1.07 on strategy); (b) stimulate private reforestation by rural and urban populations; (c) prevent further degradation of natural forests; (d) assist Fanalamanga in attracting private investment for wood- processing activities once the Mangoro plantation reaches the production stage; and (e) assist the above activities with a seed station and an applied research program. B. Proiect Design and Summary Description 3.03 Project Design. Considering the deterioration of Madagascar's capability to deal with environmental issues in general, and forestry issues in particular, and given severe financial and human resources constraints to address those issues, a restricted range of priority tasks and geographical areas has been defined and agreed with GOM. Project design addresses major institutional constraints mentioned in para. 2.08 to 2.10 by placing an emphasis on selectivity, investment in human resource development, and phased institutional development actions to allow effective absorotion of technical assistance and training. The different activities of the projects (except for the Mangoro Plantation) have been placed in existing relevant Divisions of the DEF structure in order to reinforce globally the whole DEF management capacity (see Chart 2). The project attempts to minimize its impact on recurrent expenditures by redistributing existing human resources. It also diversifies support to forestry sector activities in order to shift the current excessive PIP focus on industrial plantations (para. 2.13). The project can be viewed as a pilot project since it represents the first step of a long-term sector development program, and the experience acquired through ita specific interventions could subsequently be applied to a wider range of activities and geographical areas. 3.04 Summary Proiect Description. The project would include the following three components: (a) Forestry Department reinforcement through institutional strengthening and human resource development activities, and support to the forest management, private reforestation, integrated forest valley development, seed collection and research programs;  - 14 - (b) natural forest Protection of the Ankarafantsika, Andasibe, Lokobe, Zahamena and Tsaratanana forest reserves, creation of a now National Park in the Andasibe region and support to DEF for the management of the protected areas network; (c) reorientation of the Mantoro plantation developed under the previous two forestry projects in Madagascar primarily toward long term sawlog production. C. Geoaraihical Coverage 3.05 Most project activities will be concentrated in the regions of Antananarivo and Mahajanga. In the Antananarivo region, forest management, integrated forest valley development, and natural forest protection activities would be located in the natural forests east of the capital where the pressure on existing forests is highest, while activities supporting reforestation by urban and rural populations would be located in areas closer to urban markets, north and west of the capital. The Mangoro plantation is located along the valley of the Mangoro, some 75 kilometers east of the capital. In the Mahajanga region, project activities involving natural forest protection and manag #ent, and private reforestation are concentrated in the area between the city of Mahajanga and the southern boundary of the Ankarafantsaika forest. The integrated forest valley development program would concern seven areas in the faritanys of Toamasina and Fianarantsoa located on the eastern escarpment. Specific support to the ZCDAFARB private reforestation program and to rural forostry would be respectively located next to Antsiranana in the north of the country and in Ihosy in the faritany of Fianarantsoa. The second phase of the natural forest protection component would include the reserves of Tsaratanana and Lokobe in the north of the country, and the Zahamena reserve situated north of the Andasib4 National Park. Location of major project activities are shown on the report Map. D. Detailed Prolect Features Forestry Department Reinforcement (US$ 8.6 million) 3.06 As described in paras. 2.08 and 2.10, tho Forestry Department (DEF) has become increasingly unable to manage and develop the forestry sector. This component aims at rehabilitating the DEF by selectively strengthening its sector management capacity, developing its human resources and supporting the iMplementation of seven key programs. A summary description of these program is provided below. 3.07 Sector Management Strenathening. Sector management strengthening activities would focus ons (i) sector prograding including PIP and PEP development; (ii) program monitoring; (iii) financial managemnt; (iv) information system; and (v) forestry promotion. Project support for sector proarammint and monitorint would consist of short-term technical  - 15 - assistance, equipment and salaries for incremental staff. Following up on PIP issues identified during project appraisal, major sector programming objectives are to provide DEF, in close coordination with the DEPF, with a capacity to update and develop the forestry sector PIP, and to derive a sector PEP. Assurances ware reached during negotiations that MPAEF would submit a three year rolling sector PEP starting with the 1989-1991 cycle, and that new forestry projects above US$ 3 million would be furnished to the Association for approval. Technical assistance and equipment to improve DEF financial management would be provided to the DEF Financial Service (in the Administration Division) over the project period. Project support would focus on accounting, financial reporting, programming and budget monitoring. Particular attention would also be given to strengthening DEF cost recovery capacity. Improvements in the DEF information system are also necessary to support effective planning and monitoring. Short term technical assistance would be provided to streamline information flows within DEF, to make periodic DEF reports more action oriented, to design a training program to implement improved information flows both at headquarters and regional service levels. Forestry promotion activities would aim at developing targeted communication instruments related to specific DEF activities. Support would include short term technical assistance, equipment and would finance the formulation of focused educational material, a newsletter to transfer experience among different projects and diffuse technical innovation from abroad or from local research. Other specific promotion activities are included in the private reforestation (para. 3.15), and integrated forest valley development (para. 3.16). The head of the 'Animation' unit to be created within the Soil Conservation Division (para. 5.03) would have main responsibility for implementation of Forestry promotion. 3.08 Project activities would be complementary to institutional development measures currently underway within MPAEF in the context of the Second Agricultural Institutions Development Project (Credit 1709-NAG). The proposed Project would complete the coverage of key institutional strengthening requirements of all MPAEF's technical departments since support to the Livestock Department is provided under Credit 1211-MAG anu support to the newly created Fisheries Department is assured by a UNDP technical assistance grant. 3.09 Human resources developnent. Considering the significant human resource constraints on sector development (cf. para. 2.09), this subcomponent is conceived to increase the number and quality of professional and technical foresters both at headquarters and in the field, to update they technical knowledge and work methods of existing technical agents, and to facilitate the insertion of new technical staff into the DEF. This objective would be pursued bys (i) financing incremental staff to be gradually integrated to the DEF (Annex 12); and (ii) providing training in priority topics to both project and DEF staff (annex 6). 3.10 Staff Training. In addition to on-the-job training by the technical assistance in the context of specific project activities, a series of training courses organized by a Training Coordinator would directly support the implementation of most DEF programs included in the project. Particular emphasis is placed on developing and teaching 'social relations' techniques which are key to evolving the relationship between  - 16 - forestry agents and rural populations from a hierarchical to a collaborative mode. The establishment of new relations is indeed a determinant to the surress of participatory schemes included in several activities project. Training for each activity would include an initial intensive session followed by annual shorter follow-up sessions. Training for the natural forest protection component would be mostly on-the-job and rely on the long term technical assistants. A training course based in the new National Park in the Aidasibe area would aim at training 30 new T-level staff specialized in the management of forest reserves over the project period. The type and scope, of project related training are contained in Annex 6. In order to fill the current gap for 'ingenieurs de conception', the project would finance a limited number of post-graduate fellowships for DEF staff with strong academic credentials. Study tours are also envisaged to expose both DEF headquarters and regional managers to experience in other countries. An internship program for students of the recently reopened technical agricultural schools (para. 2.09) and for recently graduated P-level foresters has been conceived to accelerate their insertion in the DEF and their knowledge of field operutions. This program would allow forestry students and recent gradtates to be paired during the academic recess with an experienced DEF field agent to perform jointly regular service activities. 3.11 Pilot Forest management. The pilot forest management program would seek revalorisation of wood resources. Such revalorization would reflect more adequately the economic and ecologic value of wood, favor private reforestation as wood production would become a more attractive investment. In addition wood resources (thinnings and untreated stands) available in the Mangoro plantations would become more competitive for processing by private charcoalers, thus reducing pressure on the eastern natural forest as the main fuelwood resource for Antananarivo. 3.12 Initially, this component would be limited to the Antananariio wood supply area (AWSA), where the pressuAe on existing forest resources is highesi (para. 6.02), by introducing a rational forest management framework in this area. Project activities would include: (i) forest resource inventory of the AWSA; (ii) identification of major conservation areas (in coordination with natural forest protection component); (iii) study of major wood products sources and markets (in coordination with Household Energy Planning component of Energy I project); (iv) identification of major areas of agricultural encroachment on forests; (v) study of ownership patterns of eucalyptus area; (vi) elaboration of AWSA framework plan; and (vii) preparation of a detailed Forest Management Plan (FMP). It is suggested that the first FMP be done for the forests surrounding the new National Park to be created in the Andasibe region (para. 3.21). Management techniques could include: total protection, sustained exploitation, buffer plantations, controlled exploitation, conversion to agriculture and clearfelling, and aim at concentrating exploitation sites in order to reduce DEF control operating costs. Management techniques should also define required implementation measures such as incentives to eucalyptus plantation rehabilitation or regeneration, and fiscal incentives for rational forest exploitation. Support to plan implementation envisaged within the project would be limited by the DEF exploitation control capacity and would include infrastructure and equipment for control barriers and vehicles and equipment for field control teams responsible for  - 17 - monitoring forest exploitation permits. If found successful following the midterm review, it is envisaged that this forest management approach would also be implemented in the Mahajanga wood supply area. 3.13 In addition, project support would include long term technical assistance in forest management, equipment and short term technical assistance to the inventory team, short term expertise in forest revenue systems, forest legislation, natural forest management and other specific areas, and training for field staff responsible for plan implementation in AWSA. The project would also strengthen forest economics capability wi.thin the Forest Exploitation Division through short term technical assistance and long term training of a DEF staff abroad. Forest inventory activities are being defined as a complement to the national resource inventory &lso financed under Credit 1709-MAG and to the Household Energy Planning Program contained in the Energy I Project. 3.14 Adjustments in the forest revenue system, and specifically of stumpage rates, represent both an important policy instrument and an instrument to generate the resources necessary to improve the management of the sector. Assurances were given during negotiations that: (a) a calendar for definition and implementation of appropriate changes in forest revenue systems for the natural forests including possible modifications of the collection system and forest exploitation regulations; (b) the stumpage rate from DEF plantations would be revised to market rates in order to bring consistency between Fanalamanga and DEF stumpage rates; and (c) increases in stumpage rates for the natural forest would be imrlumented following recommendations of the technical assistance, DEF analysis and IDA review. Valorization of wood resources through adoption of these measures would provide further incentive for private involvement in reforestation. It could also lead to a reconsideration of new plantation investments by GOM based on the information obtained in the AWSA master plan and stumpage rates which better reflect the renewal costs of the wood resource. 3.15 Private reforestation (ZODAPARB program). The project would assist the implementation of the ZODAFARB program (para. 2.07) in five major areas located in the regions of Antananarivo (2), Mahajanga (2) and Antsiranana (1). Project support would consist of road rehabilitation and tracks to access plantation areas, nurseries, equipment for land preparation and planting, training and short term local technical assistance. Whenever possible, DEF would contract seedling production to local farmers in areas adjacent to the plantation area. DEF seedlings would be restricted for use to small farms (less than 20 ha), while larger farms would be responsible for producing their own seedlings. This would create a demand for seedlings which could be met by the development of private nurseries providing employment to the local population. Other employment opportunities would arise from firebreak and plantation maintenance requirements. The development of private nurseries requires however the abrogation of decree 4172/85 which stipulates the free cession of tree seedlings. The abrogation of this decree and the freeing of seedling price would be a condition of effectiveness. It is expected that small scale ZODAFARB participants (both local villagers and urban participants) would plant 270 ha/yr while total available land for the two ZODAFARBs in the Antananarivo area reaches 16,200 ha. Considering market  - 18 - conditions for fuelwood, the accessibility of the ZODAPARB and the investment potential in the capital city, most large scale participants are likely to come from Antananarivo. Urban collectivities would also be encouraged to participate in the ZODAFARB program. The project would finance a short promotion campaign in Antananarivo presenting benefits from ZODAFARB program. Training will be provided to DEF managers and field agents responsible for the implementation of the ZODAFARB program. Such training would not be restricted to ZODAFARB areas included in the project. Training would also be available for participants and individuals who could be hired as technicians by private participants. 3.16 Intearated Forest Valley Development. The Integrated Forest Valley Development (AIVF) program addresses the issue of slash and burn cultivation (tavy) in the eastern forest by developing a set of measures to reduce the pressure of agricultural encroachment on the remaining forests. These measures aim principally at increasing agricultural productivity of the forest valley bottoms and previously cleared slopes. The AIVF program builds on the experience acquired from the forest valley improvemen; program (AVY) since 1954, and the UNDP-financed Savoka project since 1978. The AVF program, operated by the existing DEF organization, (Soil Conservation Division) has an important rate of failure due to a dispersion of activities in too many sites, poor technical implementation by village commumities and lack of training and extension of farmers who have then abandoned the sites. The Savoka project is concentrated in two pilot regions (Faritany of Fenerive and Fianarantsoa) with precise integrated activities and close monitoring and training, but is expensive and hence is hardly sustainable. The Savoka project is finishing the construction of a third center in Mananara. The Savoka project has a separate organization within DEF. The purpose of the AIVF program would be to merge the two above approaches (AVF and Savoka) into a sustainable model within the existing DEF structure. 3.17 The program would have two phases. Phase one (years 1-3) would consist of designing the program and testing it on three existing sites, which were built in the Savoka project (Vavatenina, Kianjavato and Mananara) on the East cost. After a mid-term review specific for this program (year 3), the second phase (year 4-7) would consist of expanding the number of sites. For planning purposes, four more sites have been envisaged. During the mid- term review, the following criteria of sustainability of the model will be useds technical (crop yield, soil and forest protection), social (individual and community real participation), financial (revenues of farmers, total cost to farmers, GOM, other donors). During the mid-term review, an assessment of the effectiveness of technical assistance will be conducted. The merging of the two existing programs (AVF and Savoka) into the AIVF program in the Soil Conservation Division (para 5.03) would be necessary. Major activities would includes (i) small scale irrigation works in forest valley bottoms; (ii) agroforestry on valley slopes for soil conservation, food crops and export crops; (iii) testing of community forest management; and (iv) support to secondary products such as fruits, apiculture and fishery as supplementary sources of income and nutrition. Project support would include the construction and equipment of Technical Support Centers (TSC), small scale hydraulic works, technical assistance and training for local community participants and  - 19 - agricultural and forestry agents. The World Food Program would support the execution of collective works involved in the AIVF program. Over the project period, if phase 2 is realised with 7 TSC as planned, it is expected that 56 collectivities would be active in the AIVF program Involving the construction of 1320 ha of irrigated areas, and the plantation of 670 ha of agroforestry species, 536 ha of block tree plantation and 268 ha of fruit trees plus forest management for 2010 ha. 3.18 Seed supply and Forestry Stations. This project activity would strengthen local capacity to collect, store and provide quality seed of forest trees to support the above mentioned reforestation initiatives (para. 3.11, 3.15, 3.16). The project would finance the infrastructure and equipment of five regional seed collection centers (Fianarantsoa, Toamasina, Mahajanga, Antsiranana, Toliara). These centers were selected on the basis of their production potential and to ensure species diversity in both exotic and indigenous species. Such diversity is important to the implementation of reforestation activities. This would directly complement the ongoing Swiss bilateral activities which focuses on central seed supply quality, infrastructure and management. Agreement was reached during negotiations on a seed pricing table and pricing adjustment procedure. Two forestry stations (Angavokely In Antananarivo and Marohogo in Mahajanga) would be rehabilitated to pro,4de a physical support to training, education, forestry promotio'&, reforestation and applied research activities. Those two stations have been selected by DEF considering road access, location relative to other project activities and quality of existing infrastructure. Rehabilitation would include buildings for educational purposes, species trial site, material and equipment for station maintenance. 3.19 Research. Limited applied research would be financed under the project in direct support to specific project activities. It would focus on: (a) rehabilitation techniques for the eucalyptus plantations around Antananarivo in support to definition of forest resources management plan (cf. para. 3.07); (b) testing of multipurpose woody species beneficial to erosion control, fertilization, fodder, fuelwood at Beforona research station, in real sites (on-farm testing) and monitoring of agroforestry measures adopted in the -context of the AIVF programs; (c) species trial, natural regeneration study and agroforestry techniques testing in region of Toliari to define appropriate rural forestry approach; (d) species trials in the Marohogo forestry station (located next to Mahajanga) to identify appropriate species for possible future plantation activities, (e) continuation of the production-oriented research in the Mangoro pine plantation, but now with more emphasis on wood utilization; and (f) the regeneration and rational utilization of the natural forest. Natural Forest Protection (US$ 6.9 million) 3.20 The objective of this component is to establish an effective protection program for priority forest conservation areas. This component is based on the results of the WWF financed 'Protected Areas Program' and has been designed in coordination with major donors currently involved in conservation in Madagascar. Together, these programs would respond to priority forest conservation requirements by including Madagascar's main biotopes and covering a basic cross-section of key ecosystems (see Annex  - 20 - 3). While most activities would be focused on specific areas, support would be provided to the central DEF Nature Protection Service (SPN) in the form of training, equipment, incremental staff and imagery (aerophotography or satellite) for major forest reserves. This component would involve two phases. 3.21 Phase One. A range of protection measures would be initially applied and tested over some 150,000 ha representative of the eastern wet evergreen forest in the region of Andasibe (around the special reserve of Analamasaotra) and over some 135,000 ha of the western dry deciduous forest In the Ankarafantsika region. A new national park (10,000 ha) would be created in the Andasibe region. Measures to be tested would include imptoved demarcation methods, patrols and enforcement of boundaries, buffer plantations, promotion of tourism as a source of income to local population and development of village participatory methods supporting forest conservation. A specific forest management plan would be defined in areas adjacent to the core protection area. The project would finance housing infrastructure, training facility, visitor center and accommodations for researchers and tourists, road access, trail network, boundary delimitation, vehicles, and technical assistance. The training centre at Andasibe would serve as a training facility for all project conservation activities. The accessibility of the Andasibe reserve combined with its training and visitor facilities make it also a major focus for education and tourism. Cost recovery through tourism activities would also support the operations and maintenance of the reserves, but has not been quantified because of lack of experience in the country. 3.22 Phase Two. Depending on the results of the first phase assessed during a mid-term review (fourth project year), forest protection measures would be expanded to the reserves of Zahamena and Tsaratanana. The Zahamena reserve would include the classified forest of Andraibabe over a total area of 93,350 ha. The Tsaratanana reserve would include the Special Reserve of Manongarivo and the classified forest of Kalobenono-Antsakay covering a total area of 138,872 ha. The Zahamena reserve would represent the largest protected area in the eastern forest. During this second phase, support would be limited to basic infrastructure and equipment for forest protection and activities geared to surrounding communities. No training facilities would be provided and minimal visitor facility would be built, mostly for research purposes. The staff for each reserve would consequently be smaller than during the first pilot phase in order to reduce recurrent costs. Reorientation of the ManRoro Plantation (USS 7.1 million) 3.23 This component would complete the conversion of the Mangoro plantation from a maximum production regime, short rotation plantation intended for a large scale pulpmill to a maximum value regime, long rotation plantation for sawlog wood (para. 1.15). Project support would focus on the implementation of a sylvicultural program primarily oriented towards long term sawlog production. The plantation currently consists of some 84,000 ha of pine (71% in Pinus kesiya and 29Z in P. caribaea, P. oocarDa and P. elliottii) and 890 ha of eucalyptus. Based on current market estimates and because of the heterogeneous quality of the plantation, further afforestation would not be pursued and the  - 21 - sylvicultural program would divide the plantation into three zones with different levels of sylvicultural treatment intensity. The area under intensive treatment would corer 25,000 ha of the existing 84,000 ha. The area under semi-intensive treatment would cover 15,000 ha while the rest of the plantation would undergo minimum treatment (only in the form of fire protection). Semi-intensive treatment would essentially consist of first thinning and low pruning. Intensive treatment would, in addition, include second thinning and high pruning. Accordingly, the sylvicultural program would consist of the following major activities: (a) outstanding conventional low pruning would be completed within the first three project years followed by liberation cutting; (b) first class stands less than 10 years old would be high pruned to obtain an export quality product at 300 stems per ha; (c) overdue thinnings would be executed on stands 12 years and younger and regular thinning would be done on trees between 9 and 12 years depending on stand quality; (d) thinning of trees between 9 and 12 years of age in site classes 1, 2 and 3; and (e) second thinning would take place 5 years after the first treatment. 3.24 The following activities would be necessary to support the implementation of the sylvicultural program: (a) fire protection involving the maintenance of 950 km of firebreaks per year and the operation of the fire fighting units equipped with manned fire towers, fire-fighting brigades and vehicles, water points, aerial support, and a central fire protection unit during the dry season; (b) road maintenance and building consisting of 250 km of secondary roads and 200 km of forest tracks per year, and the yearly upgrading of 30 km of the central all-weather service road, and rehabilitation of a limited number of bridges. Building maintenance would be limited to public buildings; (c) inventory programs to provide accurate monitoring of plantation resources; (d) pasture improvement for local village communities adjacent to the perimeter which have lost grazing space to the plantation, allowing Danalamanga to maintain adequate relations with these communit4.es and mitigating the fire danger; and (e) research to monitor the impa-:t of the sylvicultural treatment. In addition to the above-mentioned activities, the project would support manatement expenditures such as supervisory personnel salary, vehicle operating expenses, and overhead on a declining basis. The Mangoro plantation component would be financially self sustained by 1994/95. Technical _Asistance and Training 3.25 Technical assistance has been included in most project activities because of the innovative nature of certain components and because of the current scarcity of foresters experienced in project supported forestry activities. The project would finance, over a period of 7 years, a total of 19.8 man-years of short term consultancies to support project implementation. In order to utilize local knowledge, local consultants have been used whenever a local consultancy capacity has been identified. Some 10.9 man-years of short term consultancies are expected to be provided by local consultants. Long-term technical assistance equivalent to 30.8 man-years (including 3.75 man-years of local expertise) would be financed in support of the following key project activities: natural forest protection (15.5), financial management (6.0), agroforestry (1.8), forest management (3.75) and training (3.75). It is expected that the 15.5 man- year long-term assistance and the 79 man-month short-term assistance, for  - 22 - the Natural Forest Protection will be provided and managed by IUCN (International Union for the Conservation of Nature). The Technical Assistance for the DEF Reinforcement component will be provided and managed technically by the operational Division/Service heads and administratively by the Project Coordinator assisted by the Financial controller who may also be assisted by a local consulting firm. It was agreed during negotiations that DEF would try to hire consultants as much as possible through consulting firms in ordez to alleviate the administrative load. A summary of technical assistance requirements is presented in Annex 4 and key terms of reference in Annex 5 (T.O.R. 1 to 10). Technical assistance would be evaluated yearly on two main criterias a) technical effectiveness and b) institution building effectiveness (do consultants transmit their experience and knowledge to local staff?). Terms of reference, for each consultant (long and short term) should mention specifically to whom local staff consultant has an institution building responsibility. The project would finance 24 man-years of overseas studies (para 3.10) at the graduate level in specialties which are in short supply in Madagascar. About 12 P- level foresters would be selected over the project period for study at the equivalent of a Master's degree. Currently identified fields of specialization include forest economics, natural reserve management, agroforestry, natural forest management, sector planning and management. Agreements would be sought during negotiations on the selection criteria for overseas study and study tour participation, on measures to ensure minimum permanence of returning degree holding forester. The project would also finance some 71 man-years of training, within Madagascar, for P- and T-level foresters of the DEF in direct support to DEF program implementation. Additional information on Project Related Training is provided in Annex 6. IV. PROJECT COST, FINANCING, PROCUREMENT AND DISBURSEMENT A. Proiect Costs 4.01 Total project cost including taxes and physical contingencies is estimated at US$ 22.6 million (FMG 47.2 billion). Foreign exchange requirements estimated at US$ 14.2 (FMG 29.6 billion) would rep esent 63% of project costs. Base cost estimates reflect June 1987 prices _..d have been adjusted to December 1987, the expected date of negotiations. Most price estimates were prepared by Fanalamanga and the mission. Physical and price contingencies over the project period are equivalent to 12% of total base costs. For the DEF components, physical contingencies have been estimated at 15% for civil works in remote areas and 101 for civil works in areas accessible from the capital city. No physical contingencies have been provided for vehicles, equipment, training and technical assistance. For the Mangoro Plantation component, physical contigencies were 10% for the pasture program, infrastructure maintenance and vehicule operations, and 5% for infrastructure investments and equipments. Price contingencies are based on expected price escalations. Price increases during the projected disbursement period are based on the following inflation rate estimates: for foreign costs, 1% for 1988/90, 2.25% for 91, and 3.5% for 1992/94; for local costs, 16.5% for 1988, 12.5% for 1989, and 101 for 1990/94. Project costs are detailed in Annex 7 and summarized in Table 4.1.  - 23 - Table 4.1: Prosect Cost Summary FMG US$ Component (million) (million) L FEX T L FEX T SFEX %BC DEF Reinforcement 3.1 7.5 10.6 2.3 5.5 7.8 71 39 Natural Forest Protection 1.9 6.4 8.3 1.4 4.7 6.1 77 30 Mangoro Plantation 5.3 3.4 8.7 3.9 2.5 6.4 39 32 Total Baseline Costs(B.C) 10.3 17.S 27.6 7.6 12.7 20.3 63 100 Physical Contingencies .3 .7 1.0 .2 .5 .7 69 4 Price Contingencies 7.0 11.6 18.6 .6 1.0 1.6 62 8 Total Project Costsa/ 17.6 29.6 47.2 8.4 14.2 22.6 63 111 at includes taxes and duties of US$2.9 million B. Prolect Financina 4.02 A financing plan by major project components is presented in Table 4.2. The proposed IDA Credit of SDR 5.1 million (US$ 7.0 million equivalent), disbursed over a period of 7 years, would be on standard terms. It would finance 36S of total project costs net of taxes, corresponding to 39Z of foreign exchange requirements and 101 of local currency requirements. The IDA Credit would finance civil works, equipment and vehicles, local training, and vehicle maintenance expenditures for the natural forest protection component (Annex 9, Table 2). The IDA credit would finance the Mangoro plantation component in accordance with Table 3 of Annex 9. The signing of a subsidiary loan agreement satisfactory to IDA between GOM and Fanalamanga would be a condition of effectiveness. 4.03 Cofinancina. Cofinancing has been sought following recelpt of an official GOM request and discussions with interested donors. Interest in cofinancing has been expressed by the Swiss Development Cooperation (SDC) and the Norwegian Ministry of Development Cooperation (NMDC). The Swiss grant funding (US $6.5 million) would cofinance foreign exchange and part of local currency requirements of the DEF reinforcement component and the training component of the Natural Forest Protection Program. Norwegian bilateral funding (US$ 2.0 million) would finance most technical assistance costs of the forest conservation component. Both sources of funding would be managed by the World Bank according to procedures to be agreed with each donor. The SDC and NMDC grants, and IDA credit would be cross effective, as a condition of effectiveness. GOM's contribution to project cost is estimated at US$ 7.1 million including US$ 3.8 million to be made through the investment budget (FNDE) and US$ 3.3 million of revenues from the sale of wood from the Mangoro plantation and cost recovery from specific project activities. During negotiations, it was agreed that a procedure would be agreed upon by February 29, 1988 in order that the cost recovery from the DVF Fainforcement activities ($ .5 million) be reinjected into the project.  - 24 - (Annex 16). GOM's contribution would finance all incremental salaries and operational costs (except vehicle maintenance) for the DEF managed components, and a share of each expenditure for the Mangoro plantation as indicated in Table 3 of Annex 9. Costs to be financed by beneficiaries in the private reforestation and integrated forest valley development components have been excluded from the financing plan; similarly, costs to be financed with food aid from the World Food Programme in support of the forest valley development program and the Mangoro plantation sylvicultural program were also excluded. Annex 9 shows the proceeds of IDA credit and of Swiss and Norway grants. Table 4.2: inancino Plan (USs m*llIon) Comonent CoM Recovery IDA SC NIDC 1 j1/ DEF Re nforcemnt 1.5 .5 C.2 6.4 8.6 Natural Forest Protection 1.8 - 8.0 .64 2.0 6.9 Mangoro Plantation 0.5 2.8 8.8 - - 7.1 etsl .8 8.8 7.6 6.5 2.0 22.6 1/ Include. texas and duties of USS 2.9 mllion paid by COM and Recovery 4.04 Provision has been made for a Project Preparation Facility (PPF) advance of US$ 950,000 refinanced under the Credit, to finance project start-up activities in DEF Reinforcement: and Natural Forest Protection components and continuity of Panalamanga operations up to June 30, 1988. Major DEF start-up activities include the set-up of a Coordination Unit including selection of financial advisor and staff, short term consultancies and equipment in farest revenue systems, forest inventory, natural reserve and site planning.  - 25 - C. Procurement Table 4.3: Procurement Methods (US$ million) Disbursement Category Procurement Method Total ICB LCB Other N.A. I. Civil workslland preparation 0.0 4.3 2.4 6.7 (-) (2.2) (1.8) (4.0) II. Vehicles and Equipment 1.7 0.9 2.6 (0.3) (0.4) (.7) III. Technical Assistance 6.3 6.3 & Training (0.7) (.7) IV. Incremental Operating 7.0 7.0 & Administrative Costs (1.6) (1.6) Total Costs 1.7 5.2 8.7 7.0 22.6 (Total IDA) (0.3) (2.6) (2.5) (1.6) (7.0) 4.05 Main categories for procurement would be goods and civil works (respectively 12X and 30S of base costs). Purchase of vehicles, equipment, tools, nursery supplies and laboratory material would be packaged whenever possible into contracts valued at US$ 50,000 and above, and would be procured in accordance with ICB procedures. The grouping should be organixed in order that, to the extent possible, there would not be more than two contracts in the same year, in particular for vehicles and major equipment. Contracts for goods for less than US$ 50,000 (aggregating to US$ 0.9 million) would be procured under LCB procedures. Civil works for DEF activities, which would be of ltmited size and distributed in different areas of the country and spread over time (average $50,000 per contract), are unlikely to attract international contractors and would be handled through LCB procedures acceptable to IDA without excluding international contractors. Road upgrading, infrastructure maintenance and sylvicultural works in the Mangoro plantation would continue to be executed by force account by Fanalamanga since they require small amounts, scattered and spread over time; moreover, this arrangement has been successful during the previous two phases. Contracts for civil works and goods, under LCB procedures will be subjected to conditions previously agreed upon concerning registration of bidders, public bid opening and local preference under Credit 1661 MAG. Technical assistance and training represent 251 of base costs. The terms of reference, qualifications, experience and terms and conditions of employment of internationally recruited experts funded by the Credit would be according to Bank guidelines. Annex 5 gives the terms of reference of key technical assistance staff. Procurement arrangements are summarised In Table 4.3 and detailed In Annex 8. 4.06 Project procurement would be managed through the DEF Coordinating Unit (para. 5.01 on Project management) and Fanalamanga. Tables indicating expected amounts covered by various procurement procedures for each managing unit are presented in Annex 8. Proposals for advertising, draft tender  - 26 - documents, bid evaluation and award proposals for civil works contracts above US$ 150,000 (LCB) and for purchases of vehicles and other goods above US$ 100,000 (ICB) would require prior IDA review before contract award. It is estimated that about 192 of total contract value would require such review. D. Disbursements 4.07 The IDA credit would be disbursed over seven years in accordance with the standard disburament profile for the Forestry Sector In Eastern Africa. The allocation of the proceeds of the IDA, Swiss, and Norway grants is presented by component in Annex 9, Tables 1, 2, and 3. Disbursements are expected to be completed by January 31, 1995. An estimated schedule of disbursements is presented in Annex 10. 4.08 Three Special Accounts (one for the Swiss GrantlDEF components, one for IDA Credit/DEF components, and one for IDA Credit/Panalamanga) would be established in freely convertable currency at the Central Bank of Madagascar to facilitate the flow of funds and reduce the number of withdrawal applications. The borrower would open the three Special Accounts and upon project effectiveness, IDA would deposit an amount equivalent to average expenditures over a four month period which is estimated at US $660,000 for the Swiss Grant/DEF account, to US$250,000 for the IDA credit/DEF account, and at US$180,000 for the IDA credit/Fanalamanga account. There will be no Special Account for the Norwegian Grant since the proceeds will pay two long-term consultants plus short-term consultants and exponses will be paid directly to the beneficiaries. The special accounts facility could be used for the PPP if necessary. IDA would be the administrator of the Swiss fund. It would review applications and instruct the Swiss National Bank to make payments. IDA would be trustee of the Norwegian fund. 4.09 Disbursements for all items would be basod on full documentation unless authorized for disbursement against statements of expenditure (SOE). Disbursements against SOE vould be used to reduce the volume of documentation between the Borrower and IDA and the length of the review process for the following expenditures: (a) civil works contracts costing less than US$50,000. (b) equipment, goods, vehicles for single items or contracts costing less than US$30,000 equivalent (c) consulting services contracts costing less than US$30,000 equivalent. (d) all force accounts and all operating costs and training. SOE's will be certified by the financial advisor with approval of the DEP Director and the Ministry of Finance for DEF components, and by the Financial Director of Fanalamanga with approval by the Ministry of Finance for the Mangoro plantation management component. The DEF financial advisor would be responsible for establishing adequate SOE procedures (para. 4.11)  - 27 - for the DEF components. The administrative and accounting capability of Fanalamanga to prepare 80 records continues to be satisfactory. Internal checks and controls are adequate and SOZe would be reviewed in detail during the annual audit (para. 4.11). Supporting documentation for SOZ would be retained by the borrower and made available for examination by IDA supervision missions. 4.10 Countergart Fundinx. The availability of local cou- erpart funds at specific times and locations is a major determinant of project implementation. Particularly in the case of tree planting which has to occur during the rainy season, availability of funds during the first trimester of the year is essential (para. 2.14). Because of MPAEF's poor investment budget implementation record to date, the submission of a proposal for the management of counterpart funds was a condition for negotiations and the Implementation of a procedure to channel GOM funds in a timely and efficient way is a condition of effectiveness. A deposit to the project account of US$130,000 equivalent at the start of each quarter, starting July 1, 1988, would be a condition of effectiveness as well. Disbursement procedures to field operations should be clearly defined for salaries, vehicle operational costs (other than vehicle maintenance), office supplies and infrastructure maintenance expenditures. The implementation of the National Forestry Fund (FFN, para. 2.14) would have a significant impact on sector financing and could affect project financing procedures. At negotiations, agreement was reached with Government that, should the FFN be Introduced during project Implementations, this would be done under term and conditions acceptable to IDA with regards to the financial management of the project. S. Accounts and Audits 4.11 The DEU, through Its Coordination Unit (CU) and the support of a financial advisor to be financed under the credit, would establish and maintain project accounts and internal control systems in accordance with acceptable accounting practices for all DEF components of the project. Manuals would be prepared to assist the implementation of control systems and of equipment control, procurement and disbursement procedures. During the previous project, Fa.alamanga has maintained a satisfactory accounting system and is currently implementing a management control system. Its Financial Department would consequently continue to maintain accounts for Mangoro plantation management activities over the project period. Assurances were obtained during negotiations that DEF and Fanalamanga would: (i) maintain records and accounts to explain all project activities; (ll) have these accounts and SOEs, if any, audited each year by external auditors acceptable to IDA; and (iii) submit these accounts to IDA no later than six months after the close of the financial year.  - 28 - V. PROJECT MANAGKMANT AND IMPLEMENTATION A. Proiect Manaaement 5.01 Ortanisation and Manatement. The Ministry of Livestock, Fisheries and Forestry (MPAhF) would have overall responsibility for Implementation of the proposed project. Responsibility for execution of project components would be delegated to DEF and Fanalamanga. A Coordination Unit (CU) would be created within DEF to support the Implementation of DEF project components (see chart 2). CU functions would be limited to functions for which DEF is not currently equipped. In order to work as much as possible within the DEF structure, the CU would not directly deal with technical issues. It would work in close collaboration with the Monitoring and Statistics Division (DASS) for monitoring and evaluation activities, and each of the DEF technical services for the technical planning and Implementation of specific project activities. The main responsibilities of the CU would includet (a) disbursement and procurement for DEF components; (b) accounts, audit and budget preparation; (c) administrative matters related to project personnel and technical assistance; (d) liaison with HPARA; (e) coordination of work program preparation and execution; (f) compilation of semestrial progress reports. The CU would be headed by a Project Coordinator (PC) (Annex 5, TOR 1) to be appointed by GOM and working under the direct supervision of the DEF Director. The DEF Director would be responsible for tho approval of work programs and budgets. The CU would also include a financial management staff composed of a financial advisor (Annex 5, TOR 2), an administrative assistant, two accountants, and secretarial staff. A Training Coordinator (para. 3.10 and Annex 5, TOR 3) would also be attached to the CU. Appointment ol the PC and recruitment of the financial advisor and the training coordinator with qualifications and experience acceptable to IDA would be a condition of disbursement for the relevant expenditure categories. Technical assistance would be attached to each of the relevant DEF technical departments. Mangoro plantation activities would be managed by Fanalamanga. B. Implementation of Prolect Components DEF Reinforcement 5.02 Prolect Staffing. The Implementation of project activities requires the assignment of DE? staff to specific functional and geographical responsibilities (para. 3.05). This would be accomplished through the implementation of a staffing plan presented in Annex 12 and the adoption of measures to create a specialized work force for the purposes of natural forest protection work. Assurances were given during negotiations on the adoption of specific position descriptions for staff assigned to the protection of natural forest who would be liberated from their 'general service' obligations. A position description for reserve agents was agreed upon at negotiations. Implementation of the staffing plan could involve reassignments within a SPIF (Service Provincial des Eaux et Forots),  - 29 - across SPEFs and recruitments within available budgetary positions. The implementation of the staffing plan including the nominal assignment of staff for each project activity and area would be a condition of disbursement for the relevant expenditures categories. Annox 12 shows the proposed staffing plan and the allocation of existing DEF staff to be reallocated and local staff financed by the project. 5 03 DEF Structure. Agreement was reached at negotiations that an 'Animation' unit would be created within the Soil and Conservation Division to support participatory initiatives (para. 3 07). This restructuration would be a condition of disbursement for the relevant expenditure categories. The current Reforestation Division would remain responsible for the development of the ZODAFARB program and for DEF plantations. Chart 2 presents the proposed DEF structure. Agreement was reached during negotiations on the adoption of such organisational structure. 5.04 Sector Management. The DEF Director would be responsible for the implementation of institutional strengthening activities in the area of sector planning and management which would be carried out by the Monitoring and Statistics Division (DASS). Notification was received during negotiations that two additional staff had already been appointed to this Division. The Project would In addition finance two incremental P-level staff to focus on sector monitoring and invastment programming (forestry PIP and PEP). Short term experts would provide direct support in the consolidation and development of the PIP and PEP, and monitoring and evaluation (para. 5.19). Close cooperation is required between the DASS and the DEPF (Direction des Etudes, Programmes et Financement) to ensure an effective integration of planning and programming procedures at the level of MPAEF. Formal periodic participation of regional staff should also be enhanced for sector planning and monitoring purposes. While the financial controller's functions would focus primarily on the project, support would be provided on a part time basis to the Financial service of DEE (para. 3.07) Additional support would be necessary in the event that the National Forestry Fund becam operational during the project period (para 2.14). A short term local expert would be hired to provide recommendations to streamline the flow of information within the DEF. Following consultations between the DEE Director, DEF managers and MPAEF, procedural changes would be adopted and implemented. Specific training activities would support this implementation at both headquarters and regional levels. Main responsibility for the implementation of forestry promotion activities would rest with the newly created 'Animation' unit within the Soil Conservation Division. 5.05 Human Resources Development The project would hire a local Training Coordinator to plan and mange the implementation of the staff training program. The Training Coordinator will be part of the Coordination Unit. Short-term consultancies would be provided at project start-up to design specific course curricula and teaching techniques. Long term technical assistance In forest management (para. 3.11), agroforestry (para. 3.15), and forest protection (para. 3.20) would monitor development and implementation of courses in conjunction with field activities. Courses would be mostly given by local personnel. Contracts could in particular be established with professors of the Agronomic School (EESSA).  - 30 - Specific training agreements could also be envisaged with the staff of the Swiss Village Reforestation Project. Candidates for study tours and higher level degrees should be submitted by the DEF Director to IDA for approval. Assurances were given during negotiations that particular attention would be given to the balance between regional and headquarters staff for study tours. The audience to local courses related to the AIVY components would include selected managers and field agents of the DVA, the DIR and the DEP. As part of its human resources development actions, the project would hire 6 P-level and 12 T-level incremental staff (see staffing plan in Annex 12). Assurances were given during negotiations that this incremental staff would be integrated into the DEF before the end of the fourth project year. 5.06 Pilot Forest Manatement. The DEF would implement the forest management activities through its Inventory and Forest Exploitation Service (SIEF) and the concerted decentralized regional services in the field. Coordination with the Envrgy I Household Energy Plan Study would be important to determine the evolution of demand including the potential of Improved charcoal burners, substitution, and fuelwood priceivolume monitoring by the Energy Unit. During preparation of the AWSA framework plan (para. 3.12), DEF should ensure that adequate consultation has been set up with different interest groups such as the Ministry of Industry and Energy, nature conservation organizations, private operators and local government representatives. Once the AWSA framework plan would be ctmpleted and regulations within a specific PMPA set, enforcement would be supported through the creation of specific field teams managed and supervised by the regional service (SPEF). Close coordination with the AIVF pilot forest management activities at the community lovel should be maintained in order to derive potential mechanisms to increase the involvement of communities in the management of local natural resources. 5.07 Private Reforestation (ZODAFARB program). The Reforestation Division of DEF would continue to be responsible for the implementation of the ZJDAFARB program. The Malagasy delegation said that l'Arrete Interministeriel n