Madagascar - Forest Management and Protection Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 15442 INFLEMENATION COMPLETION REPORT MADAGASCAR FOREST MANAGEMENT AND PROTECTION PROJECT (CREDIT 1878-MAG) MARCH 19, 1996 Agriculture and Environment Division Central Africa and Indian Ocean Department Africa Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS End of Period Figures Sources: International Financial Statistics, 1994 and February 1996 Malagasy Francs (Fmg) Year per SDR per US$ 1985 698.4 635.8 1986 941.6 769.8 1987 1751.0 1234.3 1988 2054.1 1526.4 1989 2014.0 1532.5 1990 2085.4 1465.8 1991 2621.5 1832.7 1992 2626.5 1910.2 1993 2695.8 1962.7 1994 5651.2 3871.1 Oct.1995 6857.4 4588.3 Note: the Fmg started floating on May 1, 1994. WEIGHTS AND MEASURES Metric system  FOR OFFICIAL USE ONLY ABBREVIATIONS AND ACRONYMS AIVF Amenagement integre des vallees forestieres ANAE Association nationale d'Actions environnementales ANGAP Association nationale pour la Gestion des Aires prot6g6es AVF Amenagement des vallees forestieres DD Direction des Domaines DEF Direction des Eaux et Forets DFS Deutsche Forstservice DVA Direction de la Vulgarisation agricole ESSA Ecole sup6rieure des Sciences agronomiques FANALAMANGA Fanjarian' Ala Ambatondrazaka Moramanga FAO United Nations Food and Agriculture Organization FMG Franc malagasy FOFIFA Foibem-Pirenena Momban'ny Fikarohana Ampiharina Amin'ny Fampandrosoana Ny eny Ambanivohitra (National Agricultural Research Institute) FTM Foiben-Taosarintanin'i Madagasikara (National Geographic Institute) GPF Gestion et Protection des Forets IDA International Development Association IFC International Finance Corporation KEPEM Knowledge and Effective Policies for Environment Management NEAP National Environmental Action Plan (= PAE) NOK Norwegian Kroner ONE Office national de l'Environnement ONF Office national des Forets PAE Plan d'Action environnementale (= NEAP) SDR Special Drawing Right SwF Swiss Franc UNDP United Nations Development Programme USAID United States Agency for International Development ZABA Zone d'approvisionnement en bois d'Antananarivo ZODAFARB Zone d'action en faveur de l'arbre FISCAL YEAR OF BORROWER Government of Madagascar (DEF) January 1 - December 31 FANALAMANGA July I - June 30 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. l   FOR OFFICIAL USE ONLY IMPLEMENTATION COMPLETION REPORT MADAGASCAR FORESTS MANAGEMENT AND PROTECTION PROJECT (CREDIT 1878-MAG) TABLE OF CONTENTS PREFACE ................................................................1i EVALUATION SUMMARY ................................................................1i A. Project Objectives ................................................................i B. Implementation Experience and Results ............................................................ ii C. Summary of Findings, Future Operations and Key Lessons Learned ............... iii PART I: EVALUATION FROM THE BANK'S PERSPECTIVE ..........................................1 A. Introduction ................................................................lI B. Statement and Evaluation of Objectives .............................................................2 C. Achievement of Objectives .................................................................3 D. Implementation Record and Major Factors Affecting the Project ......................4 E. Project Sustainability .................................................................6 F. Bank's Performance .................................................................7 G. Borrower's Performance .................................................................8 H. Assessment of Outcome .................................................................8 1. Future Operation .................................................................9 J. Key Lessons learned ................................................................ 10 PART II: PROJECT REVIEW FROM BORROWER'S PERSPECTIVE ........................... 12 A. Basic Information ................................................................ 12 B. DEF component ................................................................ 14 C. FANALAMANGA Component ................................................................ 17 PART III: STATISTICAL ANNEXES ................................................................ 21 Table 1: Summary of Assessments ................................................................ 21 Table 2: Related Bank Credits and Project Preparation Facilities ......................... 22 Table 3: Project Timetable ................................................................ 23 Table 4: Credit Disbursements: Cumulative Estimated and Actual ...................... 23 Table 5: Key Indicators for Project Implementation ............................................. 24 Table 6: Key Indicators for Project Operation ....................................................... 26 Table 7: Studies Included in Project ................................................................ 27 Table 8A: Project Costs ................................................................ 28 Table 8B: Project Financing ................................................................ 28 Table 9: Economic Costs and Benefits, (FANALAMANGA only)* .................... 28 Table 10: Status of Legal Covenants ................................................................ 29 Table 11: Bank Resources: Staff Inputs (expressed in staffweeks) ....................... 30 Table 12: Bank Resources: Missions ................................................................ 30 APPENDIX A: COMMENTS BY THE BORROWER ........................................................... 31 APPENDIX B: ECONOMIC ANALYSIS OF FANALAMANGA ........................................ 33 APPENDIX C: EVALUATION OF FANALAMANGA .......................................................... 38 MAP OF MADAGASCAR IBRD NO. 20657 ................................................................ 40   IMPLEMENTATION COMPLETION REPORT MADAGASCAR FORESTS MANAGEMENT AND PROTECTION PROJECT (CREDIT 1878-MAG) PREFACE 1. This is the Implementation Completion Report (ICR) for the Forests Management and Protection Project in Madagascar, for which Credit 1878-MAG in the amount of SDR 5.1 million equivalent was approved on February 16, 1988 and made effective on November 28, 1988. In November 1994, SDR 1.3 million were canceled as a result of the transfer of the DEF component from Credit 1878-MAG to Credit 2125-MAG (Environment Project). 2. The Credit was closed on November 3, 1995, earlier than the original closing date of January 31, 1996. Final disbursement took place on June 22, 1995. Upon closing, a balance of SDR 0.25 million was canceled. 3. Cofinancing for the project was provided by the Swiss Confederation (Switzerland) in the amount of SwF 10.4 million (the Swiss Contribution) and the Kingdom of Norway in the amount of NOK 13 million (the Norwegian Grant). In November 1994, SwF 7.2 million and NOK 12.6 million were canceled as a result of the transfer of the DEF component from Credit 1 878-MAG to Credit 2125-MAG (Environment Project). The Government's contribution is estimated at US$4.8 million. 4. The Preface, Evaluation Summary, Parts I and III, and the Appendixes of this ICR were prepared by Benoit Bosquet and Michel Simeon, Task Managers (AF3AE) with Connie Luff, Operations Analyst (AF3AE), and reviewed by Nils Tcheyan, Division Chief (AF3AE), and Padmanabha Hari Prasad, Operations Adviser (AF3DR). 5. The ICR is based, among other documents, on the Staff Appraisal Report, Development Credit Agreement, mid-term review, and supervision reports for the Forests Management and Protection Project, the Project Completion Reports for the First and Second Mangoro Forestry Projects (Credits 525-MAG and 1161-MAG), the Project Performance Audit Reports for the First and Second Mangoro Forestry Projects, a completion mission carried out by Guy Andriantsara (Economist, Consultant) in September-October 1995, and the findings of an evaluation of the firm FANALAMANGA by Bernard Lapeze (Forestry Specialist, Consultant) in September 1994. 6. The Borrower contributed to the ICR by preparing its own evaluation of the Project's execution (Part II) and providing comments that are included in an appendix.  I  IMPLEMENTATION COMPLETION REPORT MADAGASCAR FORESTS MANAGEMENT AND PROTECTION PROJECT (CREDIT 1878-MAG) EVALUATION SUMMARY A. Project Objectives 1. The objectives of the Project were: (a) to strengthen the capacity of the Waters and Forests Directorate (Direction des Eaux et Forets or DEF) to manage the forestry sector and carry out priority programs to implement the Borrower's National Conservation Strategy and Forestry Policy; (b) to stimulate private reforestation; (c) to prevent further degradation of natural forests; and (d) to assist FANALAMANGA in developing the Mangoro plantation to a production level which would allow sustainable and profitable exploitation. 2. To achieve these objectives, the Project consisted of three major components to be carried out by two implementing agencies. First, the forestry development reinforcement component, entrusted to DEF, included: (a) improvements in sector management by strengthening activities such as sector programming, program monitoring, financial management, information system and forestry promotion; (b) the formulation and implementation of a policy for forest exploitation regarding forest revenues and cutting permits; (c) the promotion of private reforestation programs consisting of sensitizing, training and supervision of the operators, development of nurseries, the acquisition of equipment for land preparation and planting, and road and track rehabilitation; (d) a program of integrated forest valley development activities addressing the issue of slash-and-bum cultivation, including sensitizing and supervision of the populations concerned, small-scale irrigation works in forest valley bottoms, agro-forestry for soil conservation, food and export crops, testing of community forest management, support to apiculture, fishery products and the construction and equipping of technical support centers together with staff training; (e) seed supply activities; (f) the rehabilitation of forestry stations; (g) applied research activities, mainly species trials. 3. Second, natural forest protection, also entrusted to DEF aimed at: (a) the introduction of forest protection measures in two gazetted forests (Andasibe and Ankarafantsika), including the creation of a national park at Andasibe, improved demarcation methods, usage of patrols and enforcement of boundaries, promotion of tourism, and development of village participatory methods to support forest conservation; (b) the expansion of forest protection measures to the reserves of Zahamena and Tsaratanana. 4. Third, the Project continued its support to the FANALAMANGA parastatal for the implementation of a silvicultural program, including: (a) a fire protection program; (b) road maintenance and construction of secondary roads and tracks; (c) inventory programs to monitor plantations; (d) pasture improvements for communities adjacent to the plantations; and (e) research to monitor the impact of the silvicultural program.  ii 5. The Project abandoned the pulpmill idea (cf. paras 6-8 hereafter) since it was demonstrated that neither producing the pulp locally nor exporting the wood for processing abroad would be profitable under current conditions (high transport costs from FANALAMANGA to the Toamasina ocean harbor and low world market prices in the early 1980s). 6. This Project was the third one financed by IDA in the forestry sector in Madagascar. The first one (First Mangoro Forestry Project, US$13.5 million from Loan 1065-MAG and Credit 525-MAG) extended over the period 1975-1982 and supported the establishment of new plantations in the Mangoro Valley aiming at supplying a future integrated silvo-industrial project around a pulpmill. Due to lower-than-expected yields, and despite outstanding implementation performance on the Borrower's part, the viability of the future pulpmill was questioned. 7. The second project (Second Mangoro Forestry Project, US$20.0 million from Credit 1161-MAG) extended over the period 1982-1988 and aimed at maintaining investments in the Mangoro plantations and the related operations at the same rhythm as under the first project, while carefully studying the silvo-industrial potential of the plantations. 8. The second Project's results were significantly different from the projections: (i) the plantation yields turned out to be unexpectedly low, preventing the construction of the pulpmill and forcing FANALAMANGA to look for alternative uses for its plantations (sawn wood essentially, which required a pruning and thinning program very late in the process); (ii) the studies on the potential uses of the existing plantations did not deliver all the expected investment options. As the Project Completion Report (PCR, 1991) for the second Project mentioned, long-term sustainability of the Mangoro operations was questionable. An economic rate of return of 7 percent was envisaged by the PCR in a pessimistic scenario (only part of the wood produced would find outlets) but was confirmed by the Project Performance Audit Report in 1994. 9. The above factors, combined with the Bank's reorientation from pure forestry exploitation to institution building and programming, forest exploitation control, assistance to villagers for reforestation, and expansion and protection of natural reserves, led it to include the Direction des Eaux et Forets (Waters and Forests Directorate) in its support strategy. All these lessons were integrated in setting the objectives of the subsequent project, the Forests Management and Protection Project. B. Implementation Experience and Results 10. DEF experienced great difficulty in implementing the project, because of both management and technical deficiencies. The main cause was recognized to be the excessive centralization at the Director's level, the lack of delegation to the Project coordinator, and the lack of a sector strategy. DEF proved inefficient at handling procurement diligently, thereby delaying the purchase of equipment and recruitment of consultants.  iii 11. DEF also had to struggle with deficiencies in equipment and training. DEF lacked the means to achieve a real impact on the rural populations and no synergies were developed with the other promotors of rural development. 12. Other implementation problems included the poor performance of some of the earlier coordinators, the successive resignations of financial advisers, the difficulty to implement a monitoring and evaluation system, the difficulty to integrate technical assistants and consultants, and the absence of follow-up or outright rejection of their recommendations. 13. Since the mid-term review in April 1992, the reorganization of the Project and the change in Directors at DEF, implementation and morale have improved significantly. By the time DEF's activities were transferred to the IDA-financed Environment Project in 1994, the Project was rated satisfactory for an implementation performance (the development objective rating remained mostly negative, as Table 12 in appendix indicates). All the activities initiated by the mid-term review (a forest inventory, a new human resource policy and a new forestry policy) and some new ones (the creation of new protected areas and the management of natural gazetted forests) are currently developing at a satisfactory rhythm under the umbrella of the Environment Project. 14. FANALAMANGA encountered problems in ensuring the succession of its older executives by younger ones. Nonetheless, it was capable of reorienting its activities away from the production of wood pulp (the initial destination of plantations in the mid-i 970s) and toward the maintenance of plantations to supply sawn wood, charcoal and resin. FANALAMANGA achieved most of the objectives assigned by the Project. 15. As far as results are concerned, FANALAMANGA has achieved a satisfactory percentage of its objectives, while DEF can show only few results on the ground. C. Summary of Findings, Future Operations and Key Lessons Learned 16. Sustainability. The sustainability prospects for the DEF components as they were executed until the mid-term review are negative: extension and rehabilitation methods proved too costly; rural populations participated only marginally in the Project's activities; there was little or no maintenance in reforestation perimeters; fires destroyed vast amounts of reforested land; rice fields were abandoned; dams and fish ponds were destroyed by cyclones and never repaired. After the mid-term review, these components were dropped and the Project was dramatically reorganized. Institutional reinforcement of DEF is now funded under the first phase of the Environment Project (1990-1996) and will continue during the second phase (1997-2001) with better chances of long-term impact and sustainability. 17. The Economic Rate of Return (ERR) for the FANALAMANGA component of the Project was estimated to be 55 percent at appraisal. However, due essentially to production volumes much lower than expected and to excessive personnel costs, the ERR at completion is estimated at 3 percent or less, even based on optimistic sales forecasts beyond 1995 and with all  iv the investments prior to the third Project considered as sunk costs. This is worse than the first and second Mangoro projects (see para 8 above). 18. As far as FANALAMANGA's financial sustainability is concerned, prospects have improved since the recovery of international market prices for pulpwood and woodpulp and the establishment of several saw mills that are expected to buy up the whole production of the enterprise. The self-financing capacity should become positive for the first time in 1996 and in the year 2000 self-financing should exceed financing needs. This should be considered as a positive development to attract investors if the enterprise were to be privatized. The long-term viability of the plantations, however, will be determined by the future of the enterprise. The Government of Madagascar will probably decide to divest itself to a certain extent; through selling all of the assets or through a joint venture remains to be seen. A full privatization might mean the disappearance of the plantations in the long run and the reversal of land to pastoral and agricultural uses. 19. Future Operation. The second phase of the Environment Project will give DEF the means to launch concrete natural forest management activities and transfer technology to local populations. In the course of the second phase, the improvement in DEF's institutional capacity will have to be monitored. The four main areas to be monitored are (i) the status of human resources within DEF and in its provincial and district branches (number of staff, means available, adequacy of skills/needs, etc.); (ii) the knowledge of the contents of forests; (iii) the adoption and development of the recently drafted Forest Code; and (iv) the adoption and impact of new forest management techniques based on rational exploitation and new pricing mechanisms. 20. After 20 years of financing FANALAMANGA's establishment and expansion (representing a US$30 million investment), IDA decided to stop its support. The business and market situations described above should allow the enterprise to sustain itself without foreign assistance and with minimal assistance from the Malagasy Government. Indicators to be followed in the future, which will determine the financial sustainability and viability of the enterprise, include (i) the volumes and values of the sales of sawn wood and pulpwood; (ii) the weight of personnel-related expenditures; (iii) the total area of the plantations; (iv) the yield of the plantations; and (v) the state of the infrastructure (roads, equipment, sawmills, etc.). 21. Key Lessons. The key lessons learned during the Forests Management and Protection Project include the following: (a) Institutional reinforcement may be needed before a project can be executed. This is true at the central and local levels. The central level must follow or modify a strategy and be given the human and financial means for this strategy; the local level must be given the means and skills to execute the project or monitor the project execution. In this Project, DEF's regional and sub-regional offices have by and large been left deprived of adequate means to fulfill their tasks. This shortfall has been taken into account for the preparation of the second phase of the Environment Project, which will carry on IDA assistance to DEF. In the second Environment Project, DEF will no longer manage activities directly  v but encourage the private sector and local populations to take over productive activities while limiting itself to a regulatory role. (b) The Project coordinating unit must be given enough autonomy from political pressures, especially if it is located within the Government. In the first years of this Project, the coordinator was too dependent upon the DEF Director's goodwill. (c) Procurement work must be started early, preferably during the preparation phase, so that commitments and disbursements can begin in the first year of the project. This has been particularly true for this Project: DEF was not familiar with Bank guidelines and took a long time to become operational. Specific procurement training during the preparation phase would have helped overcome this deficiency. (d) Monitoring & evaluation should be the subject of attention right from the start of the project and the M&E system should be based on systematic and reliable indicators. In this Project, the M&E function has been neglected, so that little quantified data is available to form a judgment on the Project's impact and assess its economic justification. (e) Close support from Bank staff, preferably staff resident in the country, is beneficial to project execution. The Project implementing agencies have clearly expressed their view that project implementation was facilitated by the presence of a Task Manager at the Resident Mission since October 1994. (f) Distortions may appear between Bank funded projects or agencies due to salary discrepancies, e.g. between a Government agency and a private agency. The loss of qualified personnel from DEF to ANGAP and the differences in salaries and travel allowances between these two agencies often mar their working relationship. (g) The work of consultants and technical assistants must be integrated in project activities and evaluated systematically to ensure correct transfer of technology. This proved difficult at DEF in the first years of the Project.  -  IMPLEMENTATION COMPLETION REPORT MADAGASCAR FORESTS MANAGEMENT AND PROTECTION PROJECT (CREDIT 1878-MAG) PART I: EVALUATION FROM THE BANK'S PERSPECTIVE A. Introduction 1. Madagascar's Environment. Madagascar's national area covered with primary natural forest has declined from about 25 percent of total surface in 1950 to 20 percent in 1972 and less than 15 percent today. Poverty and the low level of agricultural technology (e.g. slash-and-bum agriculture) compounded by a rapid population increase (over 12 million, rising at nearly 3 percent a year) are the main causes of natural resources degradation, including deforestation, vegetation fires, soil erosion and loss of fertility. Forest cover will disappear within 25 years if current trends are to continue. 2. The negative impact of environmental degradation on the economy is very high: the economic cost of lower agricultural productivity due to soil loss and siltation, damaged infrastructure, and the need to build new infrastructure to higher standards was estimated to equal between 5 and 15 percent of Madagascar's GNP annually in 1990. 3. Furthermore, the country is losing largely endemic species and essential ecosystems of environmental, genetic, and medical importance, thereby making Madagascar one of the world's top priorities in terms of environment and conservation. 4. The World Bank in Forestry in Madagascar. This Project is the third one financed by IDA in the forestry sector in Madagascar. The first one (First Mangoro Forestry Project, US$13.5 million from Loan 1065-MAG and Credit 525-MAG) extended over the period 1975- 1982 and supported the establishment of new plantations in the Mangoro Valley aiming at supplying a future integrated silvo-industrial project around a pulpmill. Due to lower-than- expected yields, and despite outstanding implementation performance on the Borrower's part, the viability of the future pulpmill was questioned. 5. The second one (Second Mangoro Forestry Project, US$20.0 million from Credit 1161- MAG) extended over the period 1982-1988 and aimed at maintaining investments in the Mangoro plantations and the related operations at the same rhythm as under the first project, while carefully studying the silvo-industrial potential of the plantations. 6. The second Project's results were significantly different from the projections: (i) the plantation yields turned out to be unexpectedly low, preventing the construction of the pulpmill and forcing FANALAMANGA to look for alternative uses for its plantations (sawn wood essentially, which required a pruning and thinning program very late in the game); (ii) the studies on the potential uses of the existing plantations did not deliver all the expected investment options. As the Project Completion Report (PCR, 1991) for the second Project mentioned, long-  2 Part 1. Project Implementation Assessment term sustainability of the Mangoro operations was questionable. The economic rate of return of 7 percent was envisaged by the PCR in a pessimistic scenario (only part of the wood produced would find outlets) but was confirmed by the Project Performance Audit Report in 1994. 7. The above factors, combined with the Bank's reorientation from pure forestry exploitation to institution building and programming, forest exploitation control, assistance to villagers for reforestation, and expansion and protection of natural reserves, led it to include the Direction des Eaux et Forets (Waters and Forests Directorate) in its support strategy. All these lessons were integrated in setting the objectives of the subsequent project, the Forests Management and Protection Project. 8. This Project abandoned the initial pulpmill idea since it was demonstrated that neither producing the pulp locally nor exporting the logs for processing abroad would be profitable under current conditions (high transport costs from FANALAMANGA to the Toamasina ocean harbor and low world market prices in the early 1980s). B. Statement and Evaluation of Objectives 9. The objectives of this Project became: (a) to strengthen the capacity of DEF to manage the forestry sector and carry out priority programs to implement the Borrower's National Conservation Strategy and Forestry Policy; (b) to stimulate private reforestation; (c) to prevent further degradation of natural forests; and (d) to assist FANALAMANGA in developing the Mangoro plantation to a production level which allowed sustainable and profitable exploitation. 10. To achieve these objectives, the Project comprised three major components executed by two implementing agencies. First, the forestry development reinforcement component, entrusted to DEF, included: (a) improvements in sector management strengthening activities such as sector programming, program monitoring, financial management, information system and forestry promotion; (b) the formulation and implementation of a forest exploitation policy relating to forest revenues and exploitation permits; (c) the promotion of private reforestation programs consisting of sensitizing, training and supervision of the operators, development of nurseries, acquisition of equipment for land preparation and planting, and road and track rehabilitation; (d) a program of integrated forest valley development activities addressing the issue of slash-and- burn cultivation, including sensitizing and supervision of the populations concerned, small-scale irrigation works in forest valley bottoms, agro-forestry for soil conservation, food and export crops, testing of community forest management, support to apiculture, fishery products and the construction and equipping of technical support centers and staff training; (e) seed supply activities; (f) the rehabilitation of forestry stations; (g) applied research activities, mainly species trials. 11. Second, natural forest protection, also entrusted to DEF, aimed at: (a) the introduction of forest protection measures in two gazetted forests (Andasibe and Ankarafantsika), including the creation of a national park at Andasibe, improved demarcation methods, usage of patrols and enforcement of boundaries, promotion of tourism, and development of village participatory  Part I: Project Implementation Assessment 3 methods to support forest conservation; (b) the expansion of forest protection measures to the reserves of Zahamena and Tsaratanana. 12. Third, the Project continued its support to the FANALAMANGA parastatal for the implementation of a silvicultural program, including: (a) a fire protection program; (b) road maintenance and construction of secondary roads and tracks; (c) inventory programs to monitor plantations; (d) pasture improvements for communities adjacent to the plantations; and (e) research to monitor the impact of the silvicultural program 13. The objectives of the Project were unrealistic (please refer to Section F on Bank's Performance.) C. Achievement of Objectives 14. As Table 5 illustrates, FANALAMANGA has achieved a satisfactory percentage of its objectives. However, the actual sales levels are well below forecasts. 15. DEF can show far fewer results on the ground due to poor or unrealistic project design and inadequate implementation capacity (insufficient means, poor leadership, inexperience in project management) on DEF's part: (a) a timber resource management plan was not developed for Antananarivo; (b) the frequent revision expected in forest fees did not occur; (c) populations did not effectively participate in the extension system put in place under ZODAFARB. DEF agents proved unable to exploit villagers' spontaneous reforestation initiatives; (d) the AIVF activities were never integrated into the work of the Agriculture, Livestock and Rural Engineering departments; (e) only two out of five seed production centers were rehabilitated due to the cost of the rehabilitation; (f) no forestry station was rehabilitated; (g) no sustainable and participative management model for parks and reserves was developed; (h) no training program was designed until after Project reorganization following mid-term review. 16. Significant amounts of money were invested in studies and technical assistance under the DEF components but achieved few results, the main reasons being the lack of coherence among the various interventions and the inability to integrate and capitalize on those resources.  4 Part I: Project Implementation Assessment D. Implementation Record and Major Factors Affecting the Project 17. Project Effectiveness and Closing. The Project became effective on November 28, 1988, i.e. eight months after signature. The closing date had been scheduled for January 31, 1996. Due to major problems in project implementation, the components under DEF supervision were drastically restructured at mid-term review in April 1992 and in March 1994 the decision was made to transfer these components to the Environment Project. This transfer resulted in the cancellation of SDR 1.3 million in November 1994 under Credit 1878-MAG as the Environment Project (Credit 2125-MAG) was sufficient to continue funding of these components. Strategically, this merger offered two advantages: (i) it regrouped all the environmental operations funded by IDA under a single Project; (ii) it allowed DEF to be integrated into the National Environmental Action Plan and become a likely implementing agency of the second phase of the NEAP. 18. After the merger, only the FANALAMANGA component remained active under Credit 1878-MAG. Proceeds under this component were exhausted in March 1995, leading to the decision to close the Credit early. Nonetheless, effective closing of the Project and cancellation of the undisbursed balance did not happen until November 3, 1995 due to delays on the Borrower's part in the justification of the Special Account. 19. Project Cost and Financing. The real Project cost was US$14.1 compared to the expected US$22.6 million at appraisal. IDA contribulted SDR US$4.8 million instead of US$7.0 million, Norway contributed US$0.05 million instead of US$2.0 million, and Switzerland US$5.8 million instead of US$6.5 million. The Government contributed US$4.8 million instead of the US$7.1 million expected at appraisal. 20. This difference between real and expected-at-appraisal costs is due to the low level of activities in the first years of the Project, the reorientation of the Project after mid-term review, and the cancellation of significant portions of the IDA, Norwegian and Swiss funds in 1994. 21. Disbursements. The disbursement performance remained satisfactory at around 80 percent of original appraisal projections until 1993. Disbursements then slowed down in part due to the reorganization of the DEF components following mid-term review. When the decision was made to continue the DEF components under the Environment Project, the equivalent of US$1.8 million was canceled (i.e. 25 percent of the initial allocation). FANALAMANGA, on the other hand, almost exhausted its allocation. 22. Implementation Problems. Until 1994, DEF experienced serious difficulty in implementing its parts of the Project, as much from a management as from a field activity point of view. The main cause of the difficulty was recognized to be the excessive centralization at the Director's level, the lack of delegation to the Project coordinator, and the lack of sector strategy. However, implementation improved significantly with the change in Directors in 1994. Until then, DEF proved inefficient at handling procurement diligently, thereby delaying the purchase of equipment and recruitment of consultants.  Part I: Project Implementation Assessment 5 23. DEF also had to struggle with deficiencies in equipment and training. Although it wanted to be directly in charge of operational activities, it lacked the means to achieve real impact with the rural populations. The synergies that should have been developed with the other actors of rural development also never materialized, with the exception of the research convention for specific purposes with FOFIFA. 24. Other implementation problems included the poor performance of the first coordinator, the successive resignations by financial advisers, the difficulty to implement a monitoring and evaluation system, the lack of integration of technical assistants and consultants, and the absence of follow-up or outright rejection of their recommendations. 25. The mid-term review mission in April 1992 concluded that the Project was creating an extra burden on DEF instead of reinforcing it. This capacity problem was compounded by the aging of its staff and the recruitment of some of the most dynamic members by the newly created institutions of the Environmental Action Plan, also financed by IDA under Credit 2125-MAG. 26. Since the mid-term review in April 1992, the reorganization of the Project and the change in Directors at DEF, implementation and morale have improved significantly. By the time DEF's activities were transferred to the Environment Project in 1994, the Project was rated satisfactory for implementation performance standpoint (the development objective rating remained mostly negative, as Table 12 in appendix indicates). All the activities initiated by the mid-term review (a forest inventory, a new human resource policy and a new forestry policy) and some new ones (the creation of new protected areas and the management of natural gazetted forests) are currently developing at a satisfactory rhythm under the umbrella of the IDA-financed Environment Project. 27. FANALAMANGA, on the other hand, despite some problems in ensuring the succession of the older executives by younger ones, was capable of reorienting its activities and achieving most of the objectives assigned by the Project. 28. Consultant Services. Little data is available on the impact of consultant services. As explained earlier in this report, this deficiency is due to the absence of a monitoring and evaluation system at DEF. DEF is only now installing such a system. 29. Even though this judgment cannot always be quantified, it is clear that the various consultants and technical assistants have brought little transfer of technology to DEF. As mentioned above, their work was not strategically integrated into DEF's mainstream activities and their recommendations sometimes turned out to be ill-adapted or were rejected outright. 30. Macroeconomic Environment. The hiring freeze imposed on the civil service in 1991 negatively impacted on the components of the Project under DEF supervision by preventing the hiring of new forestry engineers to replace those who reached retirement age. This problem was compounded by the general uncertainty which prevailed in 1991: the DEF personnel largely followed the general strike in the capital that lasted from May until December 1991, while insecurity hindered follow-up missions to the provinces.  6 Part I: Project Implementation Assessment E. Project Sustainability 31. The sustainability prospects for the DEF components as they were executed until mid- term review are rather grim: (a) extension and rehabilitation methods proved too costly; (b) rural populations participated only marginally in the Project's activities; (c) there was little or no maintenance in reforestation perimeters; (d) fires destroyed vast amounts of reforested land; (e) rice fields were abandoned; (f) dams and fish ponds were destroyed by cyclones and never repaired. 32. This dimension is reflected by the poor Project Development Objective Ratings (PDOR) recorded after July 1990, as Table 12 indicates. With the exception of an improvement in January 1991, PDOR were unsatisfactory. However, in December 1993, the decision was made to upgrade the Project to "2" for it was judged that the reorganization of the Project would contribute to achieving the development objectives and all measures were being taken by DEF to carry through this reorganization. This judgment was reversed in April 1994 when renewed delay in launching the three institutional studies was recorded and the PDOR fell back to "3". 33. The DEF components, as they emerged out of the mid-term review and were later transferred to the Environment Project, offer better prospects for sustainability although they are still on-going: DEF badly needed capacity building in the fields of policy formulation, development of human resources and knowledge of the potential of forests. The final results of the research in these areas will be available by the end of the first phase of the Environment Project (December 1996) and will represent the new bases for forestry work in Madagascar. Several features already stand out: disengagement of the state and transfer of responsibilities to the private sector or local populations, increased forest revenue by improved forest taxation (auction-based concession system), and decentralization of forest management from the capital to the regions. The application of these principles should help curb the degradation of natural resources, including forests. 34. FANALAMANGA's financial sustainability is still uncertain but prospects have improved since the recovery of international market prices for pulpwood and woodpulp and the establishment of several saw mills that are expected to absorb the whole production of the enterprise. The self-financing capacity should become positive for the first time in 1996 and in the year 2000 self-financing should exceed financing needs. This should be considered as a positive development to attract investors if the enterprise were to be privatized. The long-term viability of the plantations, however, will depend on the future of the enterprise. The Government of Madagascar will probably decide to divest itself to a certain extent; whether through selling all of the assets or through ajoint venture remains to be seen. A full privatization  Part I. Project Implementation Assessment 7 might mean the disappearance of the plantations in the long run and the reversal of land to pastoral and agricultural uses. F. Bank's Performance 35. Project Concept and Design. An evaluation of the Project's objectives should highlight that the absorptive and adaptive capacity of DEF was overestimated. The design of the DEF components was too broad and represented a collection of pilot activities instead of a clear set of programs with an underlying strategy. Not enough emphasis was placed on the institutional reinforcement and too much on DEF's role in promoting pilot activities. DEF's capacity to encourage and monitor, rather than control, private activities at the local level was neglected, both in terms of human skills and material means. 36. The broad goals for FANALAMANGA were correctly identified: to increase and sustain the level of production and help the firm find buyers and investors. However, the resulting sales levels were largely overestimated for the period 1988-1995. Also, as the present situation testifies, the strategy of promoting charcoal making and the production of sawn wood (for construction) and sawmill logs (for construction and furniture) was not the right one. FANALAMANGA is planning to discontinue the production of charcoal and other secondary products (fuelwood, resin) after 1995. In addition, current forecasts indicate that pulpwood will make up 80 percent of total sales from 1995 onwards, relegating sawn wood and logs to places of minor importance. 37. Supervision. As Table 11 in appendix indicates, the Project enjoyed intensive supervision on the Bank's part (over 18 staffweeks per year on average between 1988 and 1995). Yet despite this rather favorable record, the Project Development Objective Rating by and large remained unsatisfactory from January 1994 until April 1994, when the agreement was reached to close the Project by transferring the DEF component to the Environment Project (see Table 12). This points to the inability of the Bank staff to cause the Project's impact to improve. The composition of missions may shed some light on this failure. The skill mix reflects the prevalence of foresters in each mission until the mid-term review, when the Project was rated "4" on both Project Management Performance and Project Development Objectives: out of 18 mission members, 10 were foresters. In addition, the Project remained unsupervised for 8 months after the mid-term review and was rated "3" (still unsatisfactory) in December 1992. The inclusion of an economist in supervision missions at an earlier stage than the mid-term review, and closer supervision during the reorganization of the Project resulting from the mid-term review might have helped retrofit the Project. 38. The presence of a Task Manager handling all operational matters in the field from October 1994 onwards helped the continuation of work programs albeit under the umbrella of the Environment Project. As an indicator of performance, handling procurement decisions from the Resident Mission allowed the turn-around time for routine procurement issues to drop to two days on average from over 10 days prior to October 1994.  8 Part 1. Project Implementation Assessment G. Borrower's Performance 39. Implementation. The DEF components suffered from severe administrative delays in procurement (due to poor internal management and complex Government procedures). These delays were scrutinized by the mid-term review mission but became even more apparent after the DEF components were transferred to the Environment Project: DEF's contracts take a longer time in materializing due to the long series of administrative authorizations required for Government agencies. This issue was raised on several occasions, the latest one being the Bank's Country Portfolio Performance Review in May 1995. It would seem reasonable to request that Government agencies financed by IDA benefit from the same procedures as any other implementing agencies, in accordance with the recommendations of the Bank's Second Country Procurement Assessment Report for Madagascar of August 7, 1995. 40. Until the mid-term review in April 1992, DEF also registered long delays in submitting audit reports. Some audit reports did not meet the standard Bank formats, which prevented an adequate information flow to IDA on the financial accountability of Project managers. Audits formats were subsequently adapted to comply with Bank requirements and are now deemed satisfactory. 41. The Project was further harmed by frequent insufficiencies in Malagasy counterpart funding and the malfunctioning of advance mechanisms for Project activities. 42. FANALAMANGA's performance has been consistently better than DEF although the enterprise encountered problems in managing its Special Account until Project closing (e.g. replenishment requests were not filed in a timely manner, forcing the Bank to several recoveries of the Special Account; the Special Account was not justified until very late, which explains why the Project remained officially open for so long after the last disbursement). 43. Finally, the Borrower's performance would have benefited from better coordination between DEF and FANALAMANGA, both at the field (recycling of thinning waste for charcoal making to supply Antananarivo) and administrative (management of funds) levels. 44. Legal Covenants. As Table 10 in Appendix reveals, the Borrower met almost all of the Legal Covenants except those (i) on audit requirements: formal deadlines and formats were not respected during the first half of the project; and (ii) those which were affected by the reorganization of the Project after mid-term review. Nevertheless, it should be noted that a Legal Covenant may be met initially but the permanent nature of this Covenant is such that effective meeting of it means systematic compliance with it throughout the Project. For instance, Counterpart funding procedures were put in place at the beginning of the Project, in compliance with Section 6.01 (c), but subsequently there were availability problems in counterpart funding. H. Assessment of Outcome 45. FANALAMANGA. At a discount rate of 8 percent, the ERR is likely to equal 3 percent or less, compared to the 55 percent estimated at appraisal. The two main reasons for these  Part IL Project Implementation Assessment 9 significantly-lower-than-expected results are (i) heavier operating costs, especially personnel costs which represent around 30 percent of total costs; and (ii) quantities sold substantially lower than expected, due to the poor quality of wood offered (originally, trees were planted for pulp production) and poor felling techniques (felling is largely performed by unskilled contractors). 46. However, thanks to (i) the expected increase in the total value of FANALAMANGA's sales (FMG 2.6 billion in 1995 from FMG 0.9 billion in 1994 and optimistic forecasts beyond 1995), (ii) the unexpected importance of pulpwood (over 80 percent of total sales from 1995 onwards), and (iii) the favorable evolution in international market prices and world demand for pulpwood and woodpulp, the enterprise is expected to secure good revenues after 1995 (sales of FMG 9 billion per year at cruising speed compared to FMG 7 billion per year estimated at appraisal). 47. DEF. No ERR was calculated for the DEF components either at appraisal or completion. I. Future Operation 48. DEF. The decision was made in 1994 that all environmental operations should be concentrated under the Environment Project. The Environment Project is the first phase of the 15-year National Environmental Action Plan (NEAP). The second phase of the NEAP will include forestry activities aimed at the sustainable exploitation of forest ecosystems and range from the implementation of the new forestry legislation currently under finalization to the execution of management plans for gazetted forests. Institutional reinforcement of DEF remains essential but the difference will reside in the decentralized dimension. To be efficient, DEF needs to be well trained and equipped at the local level. DEF should no longer manage activities directly but encourage the private sector and local populations to take over productive activities while limiting itself to a regulatory role. 49. In the course of the second phase of the NEAP, the improvement in DEF's institutional capacity will need to be monitored. The four main areas to be monitored are (i) the status of human resources within DEF and in its provincial and district branches (number of staff, means available, adequacy of skills/needs, etc.); (ii) the knowledge of the contents of forests; (iii) the adoption and development of the recently drafted Forest Code; and (iv) the adoption and impact of new forest management techniques based on rational exploitation and new pricing mechanisms. 50. FANALAMANGA. After 20 years of financing FANALAMANGA's establishment and expansion (representing a US$30 million investment), IDA decided to stop its support. The business and market situations described above should allow the enterprise to sustain itself without foreign assistance and with minimal assistance from the Malagasy Government. 51. FANALAMANGA needs to recenter its activities and give up transformation of pine logs into finished products, carbonization and livestock production -- these activities being unprofitable. The enterprise also needs to invest heavily in a semi-mechanized felling technology. The privatization option, already discussed in the Project Performance Audit Report  10 Part I: Project Implementation Assessment for the Second Mangoro Forestry Project (1994), is feasible if the price is set right, but this would probably lead to the disappearance of the plantations in the long run. Similarly, total disengagement of the state and winding up FANALAMANGA would cause the forest to disappear because of the pressure exerted by surrounding communities. (Please refer to Appendix C.) 52. Indicators to be monitored in the future, which will determine the financial sustainability and viability of the enterprise, include (i) the volumes and values of the sales of sawn wood and pulpwood; (ii) the weight of personnel-related expenditures; (iii) the total area of the plantations; (iv) the yield of the plantations; and (v) the state of the infrastructure (roads, equipment, sawmills, etc.). J. Key Lessons learned 53. The principal lessons learned from the Project can be summarized as follows: (a) Institutional reinforcement may be needed before a project can be executed. This is true at the central and local levels. The central level must develop a strategy and be given the human and financial means for this strategy; the local level must be given the means and skills to execute the project or monitor the project execution. In this Project, DEF's regional and sub-regional offices have by and large been left deprived of adequate means to fulfill their tasks. This shortfall has been taken into account for the preparation of the second phase of the Environment Project, which will carry on IDA assistance to DEF. In the second Environment Project, DEF will no longer manage activities directly but encourage the private sector and local populations to take over productive activities while limiting itself to a regulatory role. (b) The Project coordinating unit must be given enough autonomy from political pressures, especially if it is located within the Government. In the first years of this Project, the coordinator was too dependent upon the DEF Director's goodwill. (c) Procurement work must be started early, preferably during the preparation phase, so that commitments and disbursements can begin in the first year of the project. This has been particularly true for this Project: DEF was not familiar with Bank guidelines and took a long time to become operational. Specific procurement training during the preparation phase would have helped overcome this deficiency. (d) Monitoring & evaluation should be the subject of attention right from the start of the project and the M&E system should be based on systematic and reliable indicators. In this Project, the M&E function has been neglected so that little quantified data is available to form a judgment on the Project's impact and assess its economic results. (e) Close support of Bank staff, preferably staff resident in the country, is beneficial to project execution. The Project implementing agencies have clearly expressed their view  Part I. Project Implementation Assessment 11 that project implementation was facilitated by the presence of a Task Manager at the Resident Mission since October 1994. (f) Distortions may appear between Bank funded projects or agencies due to salary discrepancies, e.g. between a Government agency and a private agency. The losses of qualified personnel from DEF to ANGAP and the differences in salaries and travel allowances between these two agencies often mar their working relationship. (g) The work of consultants and technical assistants must be integrated in project activities and evaluated systematically to ensure correct transfer of technology. This proved difficult at DEF in the first years of the Project.  IMPLEMENTATION COMPLETION REPORT MADAGASCAR FORESTS MANAGEMENT AND PROTECTION PROJECT (CREDIT 1878-MAG) PART Il: PROJECT REVIEW FROM BORROWER'S PERSPECTIVE MINISTRY OF AGRICULTURE AND RURAL DEVELOPMENT A. Basic Information Identification 1. The main objective of this institution building project was to help the Water and Forests Department (DEF) preserve ecosystems and biological species in Madagascar and assist in implementing the national conservation strategy and forestry policy, adopted respectively in 1984 and 1985. 2. Preparation of the third phase of the Mangoro Project, which has become one of the components of the Forests Management and Protection Project, was based, firstly, on the inability of parastatals to substitute for private enterprises in marketing and even in processing, as demonstrated by the World Bank's experience in providing assistance to the agriculture sector; and secondly, the questionable feasibility and rationale of the large-capacity woodpulp plant originally planned. 3. This section is based, among other documents, on the Evaluation Report (MAG 6994, 23 December 1987), the Credit Agreement (MAG 1878, 23 March 1987). The project was approved by Interministerial Decree No. 6480/88 on 28 November 1988 and the credit was ratified by Decree No. 88.024 on 24 August 1988 The project was completed in February 1995. 4. Components: (a) PHASE ONE: 1988-1992: The Forests Management and Protection Project comprised two components: (i) DEF Component a. Strengthening of DEF: Training, forest management, private reforestation, integrated planning of forest valleys, seeds and forestry stations, research, coordination of activities, b. Protection of natural forests and forest reserves: Andasibe natural reserve, Ankarafantsika natural reserve, support for nature protection service  Part II. Project Review From Borrower's Perspective 13 (ii) FANALAMANGA component (b) PHASE TWO: 1992 - August 1994 (i) DEF Component: Following two self-evaluations in February and April 1991, and the review of mid-March 1992, the project's objectives were redirected towards the following activities: a. Forestry policy b. National forestry inventory c. Human resources d. Classified forests e. Follow-up f. Research g. Coordination of activities (ii) FANALAMANGA Component (c) PHASE THREE: August 1994 - Completion The World Bank supervision missions that took place in April and June of 1994 recommended that Credit 1878-MAG be canceled and the DEF component activities be included in the First Environmental Program (PE 1) under the Environmental Action Plan (PAE) financed by Credit 2125-MAG. Only the FANALAMANGA component remained active under Credit 1878-MAG, while the remainder of the GPF project (DEF component) became an integral part of the DEF component of PEI financed by Credit 2125-MAG, with the following activities: (i) Forestry policy (ii) National forestry inventory (iii) Human resources (iv) Classified forests (v) Follow-up (vi) Research (vii) Creation of protected areas  14 Part I1: Project Review From Borrower's Perspective (viii) Coordination of activities 5. Executing Agencies: DEF and FANALAMANGA 6. Sponsoring Ministry: (a) 1989-92: Ministry of Livestock Production and Water and Forests (b) 1992-95: Ministry of Agriculture and Rural Development, subsequently Ministry of Agriculture, Livestock, Water and Forests. Financing 7. The project was cofinanced by: IDA SDR 2,200,000 Swiss cooperation SwF 10,400,000 Norway NKr 13,000,000 the Malagasy Government FMG 7,250,000 B. DEF component Project Coordination 8. For the component providing support to DEF, the project's administrative and financial management was entrusted to a coordinating unit. This choice was dictated by the fact that the Director of Water and Forests also directs the project and manages the Credit. 9. Project Implementation (a) PHASE ONE: The project is divided into two components: (i) Institutional strengthening for DEF: a. training, b. forestry management, c. private reforestation, d. integrated planning of forest valleys,  Part II. Project Review From Borrower's Perspective 15 e. seeds and forestry stations, f. forestry research and project coordination. (ii) Natural forests: a. planning of the Ankarafantsika reserve, b. planning of the Andasibe reserve, c. strengthening of the nature protection service. (iii) Shortcomings in project design included underestimating the DEF human resources problem in both qualitative and quantitative terms, and failing to take into account the freeze on recruitment in the public service and the lack of sufficient basic data on the forestry sector. (iv) Project start-up was slow because the project documents (appraisal report and credit agreement) did not plan for either the start-up phase or the activity planning and programming phase. The activities ought to have begun immediately, whereas the items needed for execution (materials and equipment) were to be obtained by tender, a process that required at least one year to complete. Project execution was poor owing to the centralization of administrative and financial management, the lack of support for decentralized services, the lack of follow-up and evaluation of project activities, and poor motivation on the part of the agents involved. Staff in the decentralized services had little interest in the project because they did not feel involved in its activities. The counterpart funds granted each year were always inadequate and never available on time. Developments in the institutional framework (creation of new PAE structures, such as ANGAP and ANAE) had rather adverse repercussions on the coordination and mobilization of DEF staff and the distribution of work. (v) Unwieldy contracting procedures both on the Malagasy side and in IDA posed a major obstacle to smooth project functioning. (vi) Still, there were several positive areas: a. some materials were acquired b. field work was done, and c. DEF engaged in project staffing efforts. (vi) In view of the foregoing, two self-evaluations took place, in February and June of 1991, and a mid-term project review was conducted in March and  16 Part IJ. Project Review From Borrower's Perspective April of 1992. After the mid-term review, DEF and IDA agreed to restate and redirect the project's objectives based on DEF's new mandate - the setting of Malagasy forestry policy, preparation of a human resources management plan, and performing a national forestry and ecological inventory. (b) PHASE TWO (i) During the second phase of the project, work took place with various partners, as follows: a. the drafting of forestry policy, with Intercooperation (Switzerland) and ESSA-For&s; b. human resources management with Orgasys-Madagascar and ONF-France; c. the national forestry and ecological inventory with DFS (Germany) and Mamokatra d. forestry research with FOFIFA; as well as e. classified forests f. the creation of protected areas (ii) The two remaining activities were managed directly by DEF, namely: a. follow-up on DEF activities, and b. coordination of activities. (iii) This second phase, lasting a year and a half, was devoted mainly to starting up the new activities, from preparation of tender documents through contract award to the procurement of materials. For disbursements, two special accounts were used: special account A for the Swiss funding and special account B for the IDA funding. At the outset of the project, the Treasury managed these special accounts through the Central Bank, but payments were delayed by the fact that the accounts were domiciled at Banque Nationale de Paris. The funds were then placed in Malagasy commercial banks and the situation improved considerably. Project accounts and direct payments are still managed by the Treasury.  Part II: Project Review From Borrower's Perspective 17 Technical assistance 10. National and particularly international technical assistance took an excessive share of the project funds. Some consulting firms experience high turnover, resulting in a lack of continuity in studies. Relations with the financial organizations 11. IDA and the Swiss cooperation supervised the project twice a year. Norway was represented by IDA in these supervision missions. During the first phase, a project officer was appointed at the resident mission, which helped in relations and project functioning. During the second phase, however, there was no one fulfilling that role and all decisions were taken in Washington, with the resident mission merely passing on correspondence. This situation made project implementation even more difficult. Since October 1994, a World Bank staff member has been posted at the resident mission at Tananarive to monitor closely the execution of the PEl Project activities. Too-frequent changes of task managers has resulted in disruptions in project management, with each individual eager to do a better job and interpreting the directives of the Bank's manuals in his own way. C. FANALAMANGA Component Project Coordination 12. The project's administrative and financial management was entrusted to the mixed public/private company FANALAMANGA, which took over from the previous manager. FANALAMANGA is in a weakened position after its restructuring, having had to terminate the contracts of a large number of staff, some of whom were well-seasoned experts who had been with the company since its founding, and firm advocates of the project. Project Objectives 13. Production of wood for various uses, from wood for energy and industrial use to sawnwood. This means, first, undertaking silvicultural work of various intensities according to the fertility of plantations; and second, increasing the rotation period from between 15 and 18 years to 25 years or more; 14. The sale of standing timber to the private sector to ensure the long-range financial sustainability of the project. This new orientation, which limits the project's role in wood production and the sale of its stands of trees, is quite frustrating for FANALAMANGA, which has planted dozens of thousands of hectares over the course of more than a decade, with acceptable results. The unstated reasons and questionable grounds for preventing the project from taking an active part in further developing its production are denying it the opportunity of boosting the value added resulting from its efforts, and thereby assuring its sustainability.  18 Part I1 Project Review From Borrower's Perspective Project implementation 15. From a technical standpoint, while observing the overall volume of work called for in the appraisal, the project opted for a balanced annual distribution of work in quantitative terms to avoid significant fluctuations in financial and human resources from one season to the next. The objectives were adequately achieved in the first years of the project. The fact that sales were well below projected figures led to a deliberate reduction in the scope of some activities, such as the second thinning and heavy pruning exercises, to a level where they had no impact on the future of the plantations. The project very soon found itself encumbered with a large quantity of cut timber littering the felling ground. This posed a fire hazard and attracted parasites, and represented an unacceptable waste of raw materials. On the sales front, results have so far been relatively disappointing, and have not achieved the year-on-year improvements projected in the appraisal report. The estimated sales increases cited in the appraisal report, though theoretically plausible, were based on certain assumptions in terms of wood availability, potential opportunities in an already tight local market under sluggish economic conditions, and the possibility of active private enterprise involvement in a sector with a virtually total lack of professionalism. Steady and increasing efforts at promotion were made, but results were slow in coming. This was quite predictable, since the only likely outlet for such a large quantity of wood, most of it of woodpulp quality, is the export market. These marketing problems cannot be resolved in the short term. Moreover, other requirements, such as transport and port infrastructure, having a direct impact on this type of activity were not properly met. Alarms have been sounded constantly since 1992 concerning the project's financial future. Those legitimate concerns have now been almost completely dispelled. Units capable of absorbing over 80 percent of the potential yearly output of sawnwood will come on stream beginning in the 1995-96 season. Also, without being unduly optimistic, negotiations under way on pulpwood promise a positive outcome in 1996-97. Management 16. Two years before the outset of this phase, with the principal future activities of the project known in general terms, training and some retraining of project staff began, and attempts were made to gain a better knowledge of the techniques involved. However, the project still faces two major problems. 17. The requirement to dismiss a large part of the staff, stipulated in the appraisal report (a reduction from 1,650 to 1,110, or one third), is creating an unprecedented erosion of morale and a deteriorating work environment. Today the decision to stop planting has been shown to be mistaken, given current demand for wood. Reducing the planting program substantially would have led to more wood being available now and would have enabled a core of skilled reforestation workers to be retained. 18. The project's far from brilliant financial position, even after the period of staff cutbacks, is causing the undesirable - and prejudicial to the project - voluntary departure of both highly skilled and intermediate employees, mainly out of fear for their employment security. The structure's adaptation to the new tasks took place very gradually. Identification of the development and production sectors from a technical and financial point of view, closely linked  Part II: Project Review From Borrower 's Perspective 19 to sales volumes, did not actually take place until the end of this phase. Delays in the release of counterpart funding each year, and the adverse effect of devaluation of the local currency in the second half of the project, have acted as constraints on the proper use of the loan. The borrower's performance in the course of project implementation has been poor compared to the first two phases of the Mangoro project, for several reasons: (a) Achieving some of the objectives assigned is contingent in part on external factors beyond its control. These include the interest of private entrepreneurs in processing the large volume of products produced in the form of charcoal and sawnwood; national economic conditions; and external market opportunities. (b) Madagascar's recent social, political and economic situation has had quite an adverse effect on project performance. 19. Despite the difficulties and unforeseen events that have arisen, a constant attempt has been made to reconcile technical imperatives with the project's at times straitened cash flow situation. 20. In the absence of a strong private sector presence, the project has had to cover too many fronts. The training and promotion centers for charcoal production and processing have inevitably become production units, despite their low capacity. The creation of a forestry operating unit to boost sales and secure the technical success of the second thinning exercise, ensuring future plantings, became essential. Although this development has not always been accepted, it has been tolerated since it has made a significant contribution to improving the project's financial position during this crucial phase. 21. Although quite recently consideration was being given to terminating the project, and despite some continuing and perhaps quite legitimate doubts, recent developments are reassuring as to the project's future, through they imply undertaking an immediate and significant effort in the coming 18 months to place the project on a more stable footing. Firm sales contracts have been signed with private companies for sawnwood; two are operational and three in the process of installation. A contract to supply 200,000 tons per year of pulpwood for chips has just been signed with a private company, and two other firms have expressed interest in the remainder. By 1996, when new units are to come on stream, the project will have accounted for the entire capacity of its plantations. Relations with the World Bank 22. The project has maintained satisfactory relations with the World Bank, despite differences in opinion from time to time as to the role the project should play to better ensure its long-term viability in view of realities prevailing in the country. Those responsible for the project hold firmly, first, that they cannot assure its profitability if it is confined to planting and selling its output in the form of standing timber; and second, that they will have to play some role in making use of its output (forestry development and processing, even if only on a professional basis) to promote development of the wood sector and to attract more private companies to the sector.  20 Part IT Project Review From Borrower's Perspective 23. Frequent supervision missions have enabled problems to be identified and addressed in a timely manner and, inter alia, the successive reallocations have enabled resources to be adapted to needs according to the situation prevailing. 24. Frequent changes in the members of supervision missions, particularly when total, do however pose problems in understanding the project's evolution, or desired evolution, and its environment, and in making optimal and timely decisions. The recent establishment of an interlocutor in the field can only enhance relations between the borrower and the financial institution. 25. More attention needs to be paid to the initial appraisal of projects, particularly in analyzing socioeconomic factors that have, or could potentially have, a significant and irreversible impact ohnimplementation of the project and its future, such as fire hazards and sources of energy available for processing.  IMPLEMELNTATION COMPLETION REPORT MADAGASCAR FORES'TS MANAGEMENT AND PROTECTION PROJECT (CREDIT 1878-MAG) PART III: STATISTICAL ANNEXES Table 1 Summary of Assessments A. Achievement of Objectives Objectives Substantial Partial Negligible Not applicable Macro policies ___ l V Sector policies l v Financial objectives _ l Institutional development _ Physical objectives _ l Poverty reduction _ Gender issues l Other social objectives _ Public sector management = _.__ ___ Private sector developmen. _ B. Project Sustainability Project Sustainability Likely Unlikely Uncertain DEF V FANALAMANGA V  22 Part III: Statistical Annexes C. Bank Performance -Bank performance Highly satisfactory Satisfactory Defic ient Appraisal J [ Supervision , [ D. Borrower Performance Borrower performance Highly satisfactory Satisfactory Deficient Preparation l Implementation l_ Covenant compliance l Operation V E. Assessment of Outcome Assessment Highly Satisfactory Satisfactory Unsatisfactory Highly Unsatisfactory [II Table 2: Related Bank Credits and Project Preparation Facilities Credit title Credit/Loan Purpose Date of signature Status Observations number Environment I Cr.2125-MAG Institution May 1, 1990 On-going Environment II building; soil under preparation conservation; biodiversity protection l Forestry I Cr. 525-MAG Plantation; December 23, 1974 Closed Ln 1065-MAG silvo-industrial complex l Forestry 11 Cr.1 161 -MAG Plantation; October 29, 1981 Closed study viability of silvo- industrial complex l Energy I Cr.1787-MAG Energy policy August 20, 1980 Closed formulation; institutional development l Energy 11 PPF 805-MAG Key reforms n.a. Negotiated Some and on conditionalities infrastructure September still to be met improvement in 12-16, 1994 energy sector  Part III: Statistical Annexes 23 Table 3: Project Timetable Step in project cycle Date planned Date actual Identification n.a. n.a. Preparation Missions n.a. March 1984 - May 1985 Pre-appraisal Mission n.a. February 1987 Appraisal Mission n.a. June 1987 Negotiations n.a. 12/14-18/1987 Staff Appraisal Report n.a. 12/23/1987 -Board presentation n.a. 02/16/1988 Signing 03/02/1988 03/23/1988 Effectiveness 11/02/1988 11/28/1988 Project completion 06/30/1995 03/31/1995 Final disbursement n.a. 06/22/1995 Closing 01/31/1996 11/03/1995 Table 4: Credit Disbursements: Cumulative Estimated and Actual (in US$ million) Year Estimate Actual Actual as % of estimated cumulative year cumulative year cumulative 1988 1.4 1.4 n.a. 1989 2.0 2.0 0.6 2.0 100% 1990 1.4 3.4 0.6 2.6 76% 1991 0.6 4.0 0.6 3.2 80% 1992 0.7 4.7 0.6 3.8 81% 1993 0.7 5.4 0.6 4.4 81% 1994 1.5 6.9 0.8 5.2 75% 1995 0.1 7.0 0 5.2 74% TOTAL 7.0 7.0 5.2 5.2* 74% * US$1.8 million canceled  24 Part 111: Statistical Annexes Table 5: Key Indicators for Project Implementation Indicators Unit Estimated Estimated Estimated at (2) as % (3) as % at appraisal at closing cruising speed of (l) of (I) ______ _____ _____ ___ __ __ (1) (2) (3) DEF Forest Management _ Forest inventoried ha 71,000 25,050 28,500 35 40 Restituted maps ha 20,000 10,000 15,000 50 75 Interpreted photos number 270 174 174 64 64 Restituted photos number 174 98 98 56 56 Satellite photos acquired ha 200,000 0 0 0 0 Forest revenues % 40 0 50 0 125 improvement Seeds Equipment regional centers number 5 2 2 40 40 Rehab. of forestry stations number 2 0 0 0 0 Seeds production tons 11.8 6.5 11.4 55 100 Local market share % 75 69 75 92 100 Research Species adaptation trials number n.a. 2,900 4,000 n.a. n.a. ZODAFARB ZODAFARB created number 5 5 5 100 100 Plantation/year number 321,000 250,000 250,000 78 78 Seedlings production/year number 540,000 416,000 416,000 77 77 Sales price/seedling FMG 2 5 5 250 250 Forest committee creation number n.a. 51 51 n.a. n.a. Opening of firebreaks km n.a. 4 4 n.a. n.a. Creation nurseries number 5 5 5 100 100 Road rehabilitation km 20 0 0 0 0 Forest tracks opened and km 10 0 0 0 0 rehabilitated AIVF Sites tested number 3 3 3 100 100 Firewood production m3/year 14,000 0 2,000 0 14 Rice production tons 3,000 126 50 4 Seedling production /year number 50,000 43,750 40,000 87 80 Villages sensitized number 56 22 n.a. 39 n.a. Areas irrigated ha 1,320 42 42 3 3 Technical support stations number 7 3 3 43 43 created Agroforestry ha 670 137 n.a. 20 n.a. Trees planted number 536 19 n.a. 4 n.a. Pisciculture basins created number n.a. 237 n.a. n.a. n.a. Tracks opened km n.a. 60 n.a. n.a. n.a. Roads maintained km n.a. 92 n.a. n.a. n.a. Traditional beehives created number n.a. 137 n.a. n.a. n.a. Naturalforest protection _ Boundary stones recovered number 49 51 51 104 104 Firebreaks opened and km 272 73 85 26 31 maintained Tracks opened and km 105 48 60 46 57 maintained  Part IM. Statistical Annexes 25 Indicators Unit Estimated Estimated Estimated at (2) as % (3) as % at appraisal at closing cruising speed of (1) of (l) l ________________________ (1) (2) (3) Air strips created km 0.5 0.5 n.a. 100 n.a. Nurseries created number 3 3 n.a. 100 n.a. Seedling production /year number 300,000 60,000 80,000 20 27 Villages sensitized number 30 21 n.a. 70 n.a. Tourist tracks maintained km 42 42 42 100 100 FANALAMANGA Planting l l_l_l_l_l Management units ha 65,500 43,180 61,000 66 93 Pruning I ha 15,000 12,225 15,000 81 100 Pruning 11 ha 25,000 11,482 25,000 50 100 Thinning ha 34,285 23,037 34,285 67 100 Replanting ha 1,366 2,040 n.a. 150 n.a. Salvage cutting ha 11,065 7,925 11,065 79 100 Permanent inventory ha 9,785 8,254 9,785 84 100 Infrastructure Road improvement km 26 17 26 65 100 Road maintenance km 416 369 416 89 100 Firebreaks km 950 636 950 67 100  26 Part III: Statistical Annexes Table 6: Key Indicators for Project Operation Estimated at Estimated at Notes I Appraisal Closing DEF l LOCAL TECHNICAL ASSISTANCE l IDA Financing 54 m/y 40 m/y Forest and park management __ 17 m/y 17 m/y Research EXTERNAL TECHNICAL 6 m/y 5.5m/y Financial Adviser ASSISTANCE LOCAL TRAINING 71 m/y 10 m/y Work planning; inventory techniques; forest exploitation; tree nurseries; microhydraulics; farmers' organizations; extension monitoring & evaluation; park guarding; forest policy TRAINING ABROAD 24 m/y 2 m/y Management; extension (8 executives trained) FANALAMANGA TECHNICAL ASSISTANCE | j IDA financing n.a. 2 m/m Completion Forestry 11 IFC n.a. 3 m/m Economic feasibility of pine log export UNDP/FAO n.a. 67 m/m Carbonization; Charcoal-making; Forestry management; Cartography and forest inventory; Marketing; Recycling of thinning waste LOCAL TRAINING Project management n.a. 565 m/d Secretarial skills; Computer science; Communications and organization Civil engineering n.a. 80 m/d Road works Maintenance n.a. 300 m/d Engines and injectors Silviculture n.a. 465 m/d Thinning and felling Product value improvement n.a. 1740 m/d Carbonization; Timber sawing and . _________________________ ____________ ____________ d ry in g TRAINING ABROAD n.a. 0 m/d = Man day m/m = Man month  Part III: Statistical Annexes 27 Table 7: Studies Included in Project Study Purpose Status Impact of study DEF _ Protected areas planning Propose management plans for 2 months - Completed Creation of 2 reserves Ankarafantsika and Andasibe Forest inventory Propose an inventory technique 2 months - Completed Method is difficult as it requires for DEF several teams, a lot of time and a very thorough training for the staff Forest fees Improve the forest fee system 3 months - Completed Increase in forest revenues Forestry sector planning Define an investment plan and a 3 months - Completed The consultant trained his methodology to prepare the counterpart, whose capacity has Public Investment Program (PIP) improved since in preparing the PIP for DEF Monitoring and Set up an M&E system for DEF 4 months - Incomplete The system proposed was Evaluation (M&E) due to project complex but helped the M&E staff restructuring in 1992 improve the capacity. Another M&E is being installed at DEF. Infornation system Reorganize the information 5 months - Complete Beginning of an improvement in system at DEF the velocity of information but disturbed by the political trouble in 1991 Microhydraulics Propose a work methodology for 7 months - Complete Rehabilitation of rice fields, the AIVF component adoption of the technique Computerization Computerize information 3 months - Incomplete The first mission report was never systems at DEF approved Community forest Consolidate and promote forest Contract signed but n.a. management management by communities canceled after mid-term review Extension Promote school-villages in Contract signed but n.a. extension canceled after mid-term review Forest management Forest management planning Contract negotiated but n.a. canceled after mid-term review Development of the Support farmers in the Contract negotiated but n.a. peripheral zones of forests productive zones around canceled after mid-tern and reserves protected areas review Sociology for forest Propose appropriate Bidding documents under n.a. exploitation methodologies for forest preparation but canceled exploitation after mid-term review Reforestation Set up a technical training plan Canceled n.a. for ZODAFARB Rural sociology for Propose appropriate Canceled n.a. ZODAFARB methodologies for ZODAFARB _ _ _ _ _ _ _~ ~~~ ~ ~ - I _ _ _ _ _ I I____ FANALAMANGA None =n.a na. n.a  28 Part Il.l Statistical Annexes Table 8A: Project Costs (in US$ million) Appraisal estimate Actual estimate Item Local Foreign Total Local Foreign Total costs costs costs costs DEF 2.3 5.5 7.8 1.0 4.8 5.8 Reinforcement Natural Forest 1.4 4.7 6.0 0.3 0.9 1.2 Protection FANALAMANGA 3.9 2.5 6.4 3.5 3.6 7.1 TOTAL 7.6 12.7 20.3 4.8 9.3 14.1 Table 8B: Project Financing (in US$ million) Appraisal estimate Actual estimate Source Local Foreign Total Local Foreign Total costs costs costs costs IDA 0.7 6.3 7.0 0.3 4.5 4.8 Swiss Cooperation 1.2 5.3 6.5 1.0 4.8 5.8 Norwegian 0.0 2.0 2.0 0.0 0.0 0.0 Cooperation Government 6.5 0.6 7.1 4.8 0.0 4.8 TOTAL 8.4 14.2 22.6 6.1 9.3 15.4 Table 9: Economic Costs and Benefits, (FANALAMANGA only)* Please refer to Appendix B Indicators Estimate at Actual at Estimate at appraisal closing cruising speed Average production of 3 sawn wood 162,000 m3/year 676 m3/year 90,000 m3/year Sales 6,968*106 FMG/year 2,628*106 9,000* 106FMG/year Economic rate of return (100% of sales attributed 55% n.a. 13.8% to third Project, @ 8% discount rate) l_l_l___ ERR (50% sales) n.a. n.a. 3% LERR (41% sales) n.a. n.a. 0% * No ERR was calculated for the DEF components at appraisal or completion  Part III: Statistical Annexes 29 Table 10: Status of Legal Covenants Covenant Topic Status Development Credit Agreement Section 2.02 (c) Opening of Special Account Met Section 3.01 (c) Cause and enable FANALAMANGA to execute Part Met C of the Project l Section 3.01 (d) Subsidiary Loan Agreement with FANALAMANGA Met Section 4.01 (a) DEF's Accounting, Reporting and Audits Met but audit report formats judged unsatisfactory in the first half of the Project Section 6.01 (b) Abrogation of Arrete No. 4 172/85 pertaining to the Met free disposition of seedlings l Section 6.01 (c) Counterpart funding procedures put in place and Met but sporadic initial deposit for Parts A and B counterpart availability problems occurred thereafter Section 6.01 (d) Conversion into equity of FMG 27.8 billion of the Met debt of FANALAMANGA to the Borrower l Section 6.02 (a) Project Agreement ratified Met Schedule 1.2 (b) (1) Appointment of project coordinator Met Schedule 1.2 (b) (2) Staffing plan of DEF Met Schedule 1.2 (b) (3) AVF and Savoka project activities merged into Met AIVF Schedule 1.2 (b) (4) Contract between DEF and DVA Met but inoperational Schedule 1.2 (b) (5) Contract between DEF and FOFIFA Met Schedule 1.2 (c) Decree creating the Andasibe National Park Met Schedule 1.2 (d) (1) Sales service established at FANALAMANGA Met Schedule 3.3 Job descriptions for natural forest program staff Met Schedule 3.4 Review of annual work plans Met but usually later than November 30 Schedule 3.5 (a) Action program for changes in the forest revenue Component canceled l ___________________________________ system after mid-term review Schedule 3.5 (b) Implementation of action program for changes in the Component canceled forest revenue system after mid-term review Schedule 3.6 New seed pricing tables Component canceled after mid-term review Schedule 3.7 Approval of IDA to the Fonds Forestier National Component canceled after mid-term review Schedule 3.9 Selection of participants for overseas study Component canceled after mid-term review Schedule 3.10 Mid-term review program Met Project Agreement Section 3.03 FANALAMANGA to take out insurance Met Section 4.01 FANALAMANGA to maintain accounting, Met reporting and audits  30 Part III: Statistical Annexes Table 11: Bank Resources: Staff Inputs (expressed in staffweeks) 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 TOTAL Preparation 1.3 2.6 8.2 38.6 45.1 1.5 97.3 Appraisal 18.1 33.1 ; = 51.2 Negotiations 14.1 . . 14.1 Supervision 13.2 20.8 11.4 26.9 34.5 11.9 14.7 11.6 145.0 ICR 2.6 _ 3.0 5.6 TOTAL 1.3 2.6 8.2 38.6 63.2 64.5 20.8 11.4 26.9 34.5 1 1.9 14.7 11.6 3.0 313.2 Table 12: Bank Resources: Missions Number Specialized staff Project Project Month/Year of skills Management Development Types of problems persons represented Performance Objective Rating Rating March 88 n.a. n.a I I June 88 n.a. n.a. I 1 Delayed effectiveness December 88 n.a. n.a. 2 1 Excessive centralization at _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ D E F March 89 2 FA,F 2 1 Procurement delays, Accounting and budgeting at ____I_I_I_ DEF November 89 1 FA 2 2 | July 90 6 FA,F,F,F,F,OO 3 3 | I January 91 4 FA,F,F,F 2 2 August91 n.a. n.a. 2 3 December91 2 F,YP 3 3 March-April 92 3 F,AE,IB 4 4 DEF: Centralization, (Mid-term Bureaucratization, Lack of Review) strategy, Lack of priorities, Poor level of personnel, Low absorptive capacity, Lack of counterpart funding FANALAMANGA: lack of | aggressive marketing December 92 1 AE 3 3 Commitment and implementing capacityof DEF April 93 1 AE 3 3 September 93 1 AE 2 3 1 11 December 93 1 AE 2 2 _ 11 April 94 2 AE,OA 2 3 1 11 June 94 3 AE,YP,OA 2 U FA= Financial Analyst YP= Young Professional 00= Operations Officer F= Forester AE= Agricultural Economist IB= Institution Building Specialist OA= Operations Analyst  APPENDIX A: Comments by the Borrower Introductory note: This is the English translation of the November 20, 1995 letter of the Director of the Waters and Forests Department (DEF), Henri FINOANA, to the World Bank presenting the Borrower's reaction to the draft ICR. Comments on the draft Implementation Completion Report for the Forests Protection and Management Project: Reminder on the situation in which the Project was launched: a) This was the first project with external funding ever managed by DEF. b) DEF had to become familiar with the administrative and financial procedures of the World Bank and the Malagasy Government, in particular with regard to procurement. As a consequence, all the goods which had to be procured (vehicles, buildings, miscellaneous equipment, consulting services) were not acquired until the mid-term review. c) In the interim period, DEF had to work with the available material and human means. d) The second half of the period leading to the mid-term review was marred by a general strike which lasted through the end of 1991. The above situation explains why DEF has not been able to achieve the results expected initially. Institutional support did not really start until after mid-term review. Thanks to the experience accumulated during the first half and above all since mid-term review, the Forests Protection and Management Project has acquired the foundations on which to build better project management: a) DEF has now acquired some experience in project management. b) DEF has acquired some capital goods. c) DEF has benefited from the studies carried out since mid-term review. d) Support from the World Bank Resident Mission benefits the whole Environment Project (EPI).  32 Appendix A e) DEF's integration in the EPI has alleviated the burden of public procurement in accordance with Malagasy Government rules (DEF no longer relies on the Commission d'Appel d'Offres nor the Commission Centrale des Marches). f) The opening of Special Accounts with local primary banks has facilitated the use of fimds.  APPENDIX B: Economic Analysis of FANALAMANGA Tables 13-15 hereafter calculate the Economic Rate of Return (ERR) for the FANALAMANGA Mangoro plantation under three scenarios, each time with a discount rate of 8 percent. The first scenario (Table 13) assumes that all the sales are attributable to the present third Project, i.e. the economic value is the same as the financial value. The second scenario (Table 14) assumes that the economic value of the sales is only 50 percent of the financial value. The third scenario (Table 15) tests the level of economic benefits that would make the ERR equal to zero. Data were provided by Mr. Bernard Lapeze (Forestry Specialist, Consultant) and Mr. Guy Andriantsara (Economist, Consultant). First scenario (Table 13) (a) Costs comprise the investments (including replacements) plus operating costs and working capital. Most capital goods being imported and subject to taxes of around 40 percent, the investments presented here equal 60 percent of the nominal values provided by FANALAMANGA. 30 percent of the operating costs being made up of personnel charges, the opportunity cost of unskilled labor in the Moramanga region estimated at 51 percent at appraisal was applied to 30 percent of the operating cost figures provided by FANALAMANGA. (b) Financial benefits equal total sales times the contract price per cubic meter of sawn wood (FMG 36,000 from 1995 onwards) and pulpwood (around FMG 10,000 from 1995 onwards) sold by FANALAMANGA to the various sawmills. These prices are in line with the reference price for sawn wood calculated by Bernard Lapeze in his evaluation of FANALAMANGA: A cubic meter of standing pine wood should be bought from FANALAMANGA at around FMG 30,000 (at an exchange rate of FMG 700 per French Franc) if it is to be sold at the average price of FF 930 to a European factory and transformed into pallets or packing crates. This price of FMG 30,000 takes all treatment and handling costs and a 15 percent profit margin for the sawmill into account. (c) All prices are adjusted for inflation and converted into millions of 1995 FMG (source: consumer price index for Madagascar, International Monetary Fund, Intemational Financial Statistics). Under this scenario, the ERR equals 14 percent compared to the 55 percent estimated at appraisal. The two main reasons for these significantly lower results are: (i) heavier operating costs, especially personnel costs which represent around 30 percent of total costs; and (ii) substantially lower sales volumes than expected at appraisal. These lower-than-expected volumes (e.g. 90,000 m of sawn wood per year at cruising speed instead of 162,000 m3 per year expected at appraisal) are due to the poor quality of wood offered (originally, trees were planted for pulp production), poor felling techniques (felling is largely performed by unskilled contractors), and forest fires.  34 Appendix B Nevertheless, from a financial perspective, thanks to (i) the expected increase in FANALAMANGA's sales (FMG 2.6 billion in 1995 from FMG 0.9 billion in 1994 and optimistic forecasts beyond 1995), (ii) the increasing importance of pulpwood (over 80 percent of total sales from 1995 onwards), and (iii) the favorable evolution in international market prices and world demand for pulpwood, the enterprise seems viable. Table 13 assumes that all the benefits can be attributed to the present third Project while the investment and operating costs incurred during the first two Projects in establishing the plantations are considered as sunk costs. This unrealistic assumption is tantamount to saying that without the present third Project, the plantations would vanish without producing any benefits whatsoever, totally destroyed by fires or cyclones. Second scenario (Table 14) Although it is difficult to quantify the incremental economic benefits accruing to the third Project alone, Table 14 reflects a more likely hypothesis: only 50 percent of the benefits accrue to this Project. In other words, 50 percent of the benefits would be foregone without the third Project: 50 percent of the forest would disappear while 50 percent would still be s6ld by the enterprise or used by local communities for alternative purposes such as construction poles and charcoal making. In these conditions, the ERR of the third Project drops to 3 percent. Third scenario (Table 15) Table 15 presents a "break-even" analysis of the reduction of economic benefits which would result in the ERR of the third Project becoming negative. For this to occur, the Project would have to be responsible for no more than 41 percent of the total benefits, i.e. 59 percent of the total benefits would still be collected by society by the year 2012 through sales of wood or alternative usages, even in the absence of the third Project. This scenario is not unrealistic. If indeed the third Project allowed no more than to conserve and sell 40 percent of the plantations and if 60 percent of the benefits resulting from the sale or alternative usage of the plantations would have accrued to society anyway without the third Project, even attributing all the costs related to establishing the plantations to the two prior Projects, the third Project has a negative impact. Based on an ERR of 3 percent or less, the incremental investment in the FANALAMANGA Mangoro plantation turns out to be even less profitable than during the first and second Projects, which together achieved an ERR of 7 percent (confirmed by the Project Performance Audit Report for the second Project in 1994). The ERR for the three Bank-funded investments in the Mangoro plantation together must therefore be negligible, if not negative.  Appendix B 35 Table 13 Calculation of the NPV and ERR for the FANALAMANGA component (Third Project only) (100% of benefits accruing, in millions of 1995 FMG) Year Gross Benef Invest Op Costs Wrk Cap Costs Net Benef Disc NB (1) (2) (3) (4) (5)=(2)+(3)+(4) (B)=(1)-(5) 1 1988 212 -3,545 -2,371 0 -5,916 -5,703 -5,281 2 1989 567 -992 -2,511 0 -3,503 -2,937- -2,518 3 1990 704 -1,255 -2,473 0 -3,728 -3,024 -2,401 4 1991 829 -1,431 -2,670 0 -4,101 -3,272 -2,405 5 1992 831 -923 -3,395 0 -4,319 -3,488 -2,374 6 1993 1,298 -680 -1,747 0 -2,427 -1,129 -711 7 1994 1,429 -594 -2,161 0 -2,755 -1,326 -774 8 1995 2,628 -822 -1,897 -765 -3,484 -856 -463 9 1996 5,640 0 -2,763 -418 -3,181 2,459 1,230 10 1997 8,136 -900 -2,912 -54 -3,866 4,270 1,978 11 1998 8,388 -990 -3,011 -24 -4,025 4,363 1,871 12 1999 9,000 -1,200 -3,011 0 -4,211 4,789 1,902 13 2000 9,324 0 -3,011 0 -3,011 6,313 2,321 14 2001 9,681 0 -3,011 0 -3,011 6,670 2,271 15 2002 10,073 0 -3,011 0 -3,011 7,062 2,226 16 2003 10,504 -300 -3,011 0 -3,311 7,193 2,100 17 2004 10,978 0 -3,011 0 -3,011 7,967 2,153 18 2005 11,500 0 -3,011 0 -3,011 8,489 2,124 19 2006 12,074 0 -3,011 0 -3,011 9,063 2,100 20 2007 12,705 0 -3,011 0 -3,011 9,694 2,080 21 2008 13,400 0 -3,011 0 -3,011 10,389 2,064 22 2009 14,164 -300 -3,011 0 -3,311 10,853 1,996 23 2010 15,004 0 -3,011 0 -3,011 11,993 2,043 24 2011 15,929 0 -3,011 0 -3,011 12,918 2,037 25 2012 16,946 0 -3,011 1,261 -1,750 15,196 2,219 Total 201,943 -13,933 -70,066 0 -83,999 117,944 17,789 NPV 17,789 ERR 13.84% d 0.08  36 Appendix B Table 14 Calculation of the NPV and ERR for the FANALAMANGA component (Third Project only) (50% of benefits accruing, in millions of 1995 FMG) Year Gross Benef Invest Op Costs Wrk Cap Costs Net Benef Disc NB (1) (2) (3) (4) (5)=(2)+(3)+(4) (6)=(1)-(5) 1 1988 106 -3,545 -2,371 0 -5,916 -5,809 -5,379 2 1989 283 -992 -2,511 0 -3,503 -3,220 -2,761 3 1990 352 -1,255 -2,473 0 -3,728 -3,376 -2,680 4 1991 414 -1,431 -2,670 0 -4,101 -3,687 -2,710 5 1992 415 -923 -3,395 0 -4,319 -3,903 -2,656 6 1993 649 -680 -1,747 0 -2,427 -1,778 -1,120 7 1994 715 -594 -2,161 0 -2,755 -2,041 -1,191 8 1995 1,314 -822 -1,897 -765 -3,484 -2,170 -1,173 9 1996 2,820 0 -2,763 -418 -3,181 -361 -181 10 1997 4,068 -900 -2,912 -54 -3,866 202 94 11 1998 4,194 -990 -3,011 -24 -4,025 169 72 12 1999 4,500 -1,200 -3,011 0 -4,211 289 115 13 2000 4,662 0 -3,011 0 -3,011 1,651 607 14 2001 4,841 0 -3,011 0 -3,011 1,829 623 15 2002 5,037 0 -3,011 0 -3,011 2,025 638 16 2003 5,252 -300 -3,011 0 -3,311 1,941 567 17 2004 5,489 0 -3,011 0 -3,011 2,478 670 18 2005 5,750 0 -3,011 0 -3,011 2,739 685 19 2006 6,037 0 -3,011 0 -3,011 3,026 701 20 2007 6,353 0 -3,011 0 -3,011 3,341 717 21 2008 6,700 0 -3,011 0 -3,011 3,689 733 22 2009 7,082 -300 -3,011 0 -3,311 3,771 694 23 2010 7,502 0 -3,011 0 -3,011 4,491 765 24 2011 7,965 0 -3,011 0 -3,011 4,953 781 25 2012 8,473 0 -3,011 1,261 -1,750 6,723 982 Total 100,972 -13,933 -70,066 0 -83,999 16,972 -10.407 NPV -10,407 ERR 3.09% d 0.08  Appendix B 37 Table 15 Calculation of the NPV and ERR for the FANALAMANGA component (Third Project only) (41% of benefits accruing, in millions of 1995 FMG) Year Gross Benef Invest Op Costs Wrk Cap Costs Net Benef Disc NB (1) (2) (3) (4) (5)=(2)'(3)+(4) (6)=(1)-(5) 1 1988 87 -3,545 -2,371 0 -5,916 -5,828 -5,397 2 1989 232 -992 -2,511 0 -3,503 -3,271 -2,804 3 1990 289 -1,255 -2,473 0 -3,728 -3,440 -2,731 4 1991 340 -1,431 -2,670 0 -4,101 -3,761 -2,765 5 1992 341 -923 -3,395 0 -4,319 -3,978 -2,707 6 1993 532 -680 -1,747 0 -2,427 -1,895 -1,194 7 1994 586 -594 -2,161 0 -2,755 -2,169 -1,266 8 1995 1,077 -822 -1,897 -765 -3,484 -2,407 -1,300 9 1996 2,312 0 -2,763 -418 -3,181 -869 -434 10 1997 3,336 -900 -2,912 -54 -3,866 -530 -246 11 1998 3,439 -990 -3,011 -24 -4,025 -586 -251 12 1999 3,690 -1,200 -3,011 0 -4,211 -521 -207 13 2000 3,823 0 -3,011 0 -3,011 812 298 14 2001 3,969 0 -3,011 0 -3,011 958 326 15 2002 4,130 0 -3,011 0 -3,011 1,119 353 16 2003 4,307 -300 -3,011 0 -3,311 996 291 17 2004 4,501 0 -3,011 0 -3,011 1,490 403 18 2005 4,715 0 -3,011 0 -3,011 1,704 426 19 2006 4,950 0 -3,011 0 -3,011 1,939 449 20 2007 5,209 0 -3,011 0 -3,011 2,198 472 21 2008 5,494 0 -3,011 0 -3,011 2,483 493 22 2009 5,807 -300 -3,011 0 -3,311 2,496 459 23 2010 6,152 0 -3,011 0 -3,011 3,141 535 24 2011 6,531 0 -3,011 0 -3,011 3,520 555 25 2012 6,948 0 -3,011 1,261 -1,750 5,198 759 Total 82,797 -13,933 -70,066 0 -83,999 -1,203 -15.483 NPV -15.483 ERR -0.25% d 0.08  APPENDIX C: Evaluation of FANALAMANGA Introductory note: These are the salient points of the report written by Mr. Bernard Lapeze (Forestry Consultant) at the outcome of his September 1994 mission. His mandate was to assess the potential for FANALAMANGA to develop into a self-financing enterprise and make recommendations to this end before the end of 20 years of World Bank support. Plantations: pine tree plantations are of mediocre quality (slow growth and small dimensions). The oldest plantations require urgent exploitation. 59,000 ha out of 84,000 ha can be exploited. Roads: the road network is well developed and allows access to all plots. However, the quality of roads must be improved to facilitate exploitation. Bridges: they are in more or less good state and need to be reconstructed every 8-10 years. Buildings: too many of them; they cannot all be maintained properly. Equipment: in more or less good state. Financial situation: FANALAMANGA has only been able to survive thanks to aid it cannot pay back. Operating expenses (excluding inventories) amount to FMG 3 billion, of which FMG 930 million are personnel costs. Operations: Timber exploitation is wholly manual, and ensured by either FANALAMANGA staff or small traders contracting out the actual exploitation to unqualified workers. Exploitation is never systematic and only the biggest trees are logged in a creaming fashion. Forests are impoverished since only the small trees are left standing. The most realistic solution is to adopt a semi-mechanized operations strategy: labor intensive but heightened productivity and more rational management. This will represent a significant investment which FANALAMANGA would need to undertake soon (FMG 1.7 billion over 5 years). Wood transformation: Three sawmills (PINE EXPORT, MORAMANGA, FANALAMANGA) have been established and three others are expected to settle down (ARCHIMBAUD, MARQUE, ZACHARIE). Wood markets: Local market: very thin for coniferous trees. Export markets: these are the only real outlets for the above sawmills. But the pine quality only allows to market low added value sawn woods. The most important customers will probably be found in neighboring Reunion, Mauritius and the Comoros. However, these countries cannot absorb the total production of the Moramanga region sawmills. More remote possibilities exist in Spain, Italy, Egypt, Israel, Saudi Arabia and Japan. Freight costs will determine whether exports to these countries are feasible.  Appendix C 39 As for pulpwood and sawmill waste products, which represent increasingly important volumes, these can only be consumed locally for technical and economic reasons. What future for FANALAMANGA? FANALAMANGA's survival depends on the industries which will settle down and purchase its wood. At the time being, none of them are working up to their capacity. Since 1984, the change in objectives has forced FANALAMANGA to look for outlets other than pulpwood (sawn wood, charcoal and resin harvest) and to change its silvicultural approach. FANALAMANGA's staff have adapted perfectly to these new techniques. Transformation has also been promoted to recycle thinning products. Charcoal is good in that it cleans up the forest after thinnings but its traditional exploitation is not profitable . Winding up FANALAMANGA would cause the forest to disappear in the medium run. - All activities related to wood transformation, carbonization and livestock production must be given up and, if possible, sold. Staff must be cut to 300 people maximum (from 1,130 currently) and assigned to the field. If the decision was made to stop FANALAMANGA's operations, the firm could be privatized. Price would determine the demand. However, selling the forest in blocks would lead to its disappearance in the long run.   A2- 4A