FINAL REPORT Final Evaluation Report: Zimbabwe Natural Resources Management Program - USAID/Zimbabwe Strategic Objective No.1. CAMPFIRE Communal Areas Management Programme for Indigenous Resources Brian Child Brian Jones David Mazambani Andrew Mlalazi Hasan Moinuddin 29 July 2003 ----------- --- NR1\\I'1 INAI .RI:I'OR I 2').11 'I , 2()(J, TABLE OF CONTENTS EXECUTIVE SUMMARY--- ---- - ----- - ---------------;,---- I. INTRODUCTION 18 A. Pl )RI'O'; I: OI' rill: EV"I. LJ ATiON 18 B. HISTORICAL BACKGROUND or CAMPFIRE 19 1. Phase I: Initiating CAMPFIRE 19 2. Phase II Funding, Programme Diversification, CAMPFIRE Association and RDCs 21 3. Phase III: Winding down of External Support 23 (" \1ACRO- FCONOMIC CO\: DIl'1 0-"': S TO CAMPFIRE 2-1 n. I.E( i" l \ '< nPOIICYi-RAI'.II·W()I{K 27 I:. USAID SI IPPORT (NRMP) TO C A:vtPf-"!RE 29 1. NRMP I (1989-1994) 29 2. NRMP II Project Design: Goal, Purpose and Outputs 30 3. Direct and Indirect Project Beneficiaries 31 4. NRMP II Implementation Strategy prior to the 1998 Mid-Term Evaluation 31 5. Revised Implementation Strategy after the 1998 Mid-Term Evaluation 31 6. Responsiveness of NRMP 1/ to Recommendations of the Mid-Term Evaluation 33 II. INVESTMENT INFLOWS UNDER NRMP I AND II 43 III. IMPACT ACHIEVED AND CHALLENGES FACED 50 A . CONSERVATiON 50 1. Awareness 50 2. Protecting Landscapes and Habitats 52 3. Management of Large Mammals 55 4. Habitat Management 58 B. ECONOMIC BENEFIT 59 C. DIVERSIFICATION 64 D. D!:VOI.trIION 69 E. CAPAC! ry Bl JlLDING AT SlJl3-DISTRIC r, DISTRICT ANI> NATIONAL LEVELS 76 1. Institutional Management Capacity 76 2. Wildlife Management Capacity 82 3. Programmatic Monitoring and Adaptive Management 85 F. GOVI·.I{?\:\N( 'I- 89 G T ECIINOI ()(ill:S AND INNOVATION 91 H NA·IION,\I . RI :liIONAI. ANI) INTERN,\ IIONAI IMPACT 91 IV. ASSESSMENT OF PROJECT DESIGN, DELIVERY AND COORDINATION 93 A . TilE CAMPFIRE DEVELOPMENT Fl iNn 93 B . ROLl: 01· 1111: CAMPFIRE ASSOCIA rlON 104 C. ROLE OF rill' I NSTiTUTIONAL CON I R:\( 'TOR (DEVELOPMENT ASSOCIATES) 107 D. ROLl' OF llSAID, HARARE OFFICE 110 E. PRO.ll:(, I MODl.1. AND PROJECT D I'SlliN I I I V. PERFORMANCE OF COLLABORATIVE PARTNERS AND SUPPORT AGENCIES 114 A MLGPW;":!-I (PREV IOl JSI Y KNOWN AS MLGRUD) 114 n. DN PWLM 115 C DNR 118 D . FORESTR Y COMMISSION I 19 E. CASS 121 F. WWF 123 G. ACTION MAGAZINE AND ART 125 H. ZIMTRUST 126 I. SA FIRE 128 VI. IMPACT OF EXTERNAL FACTORS ON PROJECT (NRMP) PERFORMANCE 130 VII. SUSTAINABILITY OF NRMP INTERVENTIONS 132 A . SUSTAINAIIILlTY OF CAMPFIRE D t'VEI OPMENT FUND B. SUSTAINABLE FUNDING OF THE CAMPFIRE ASSOCIATION C. SUSTAINAI3ILlTY OF COMMUNITY PROJECTS FUNDED UNDER CDF VIII. SUMMARY OF FINDINGS: POSITIVE IMPACT AND WEAKNESSES A . POSITIVE IMPACT . B. WEAKNESSES OF NRMP II IX. LESSONS LEARNED X. RECOMMENDATIONS AND FUTURE OUTLOOK NRMPFINAI.REPORT 2 132 133 138 140 140 142 144 146 LIST OF FIGURES Figure l' Zimbabwe Population and Employment .... ...... .... .. .. .. .. ..................................... ....................... 24 Figure 2. Zimbabwe Gross Domestic Product (USD billion) .. ..... ......... .... .. .... ............ . ...... .......... .......... 25 FIgure 3: Zimbabwe tourism : Regional and International Arrivals .............. .. .............. .. .. .. .. ........................ 26 Figure 4: The Number of Animals Hunted in Zimbabwe, 1984 to 2000 ................ .. .. ... .. ......... ... ................ 26 Figure 6: Map of CAMPFIRE "Producer Wards" in 1996 ................ ...................... .. ............. ....................... 52 Figure 7: Trends in Quotas set for Big-Game and Plains-Game in CAMPFIRE Areas ........................... .. ... 55 Figure 8: Elephant Population Estimates in CAMPFIRE Areas ............................. .. .. ...................... ........... . 56 Figure 9: Trends in Average Trophy Size of Elephants on State and Communal Areas ... .................... ...... 56 Figure 10: Trends in Buffalo Populations in Communal Areas ................................... ...... .. .................... ....... 57 Figure 11 : Trends in Buffalo Trophy Quality on Communal and State Areas ................................. .............. 57 Figure 12: Lion and Leopard trophy Quality Trends in Zimbabwe (1989-2002) .... .. .. ... ... .. ... .. .. ...... ............. .. 58 Figure 13' Direct Income to CAMPFIRE .. .. .. ............................................................ ............. .. .. ... .. .... ........ 59 Figure 14. Factors Contributing to Increasing Income: Improving Marketing and Increasing Wildlife Populations ................ ............ ... .... .. ............ ....... ............ ....... ......................... ................... ...... .... 59 Figure 15: Summary of Early Progress in Fiscal Devolution ................ ..... .......... ...................... .................... 62 Figure 16: Summary of Allocation of CAMPFIRE revenues to RDCs and Producer Communities (USD) (1989-2001 ) .................. ......... .. .............................. .. .................................................................... 62 Figure 17. Summary of Allocation of CAMPFIRE revenues to RDCs and Producer Communities (1989- 2001) .................. .. .. ............................... .. ................ .. .................................................................. 62 Figure 17: Summary of Allocation of CAMPFIRE revenues to RDCs and Producer Communities (1989- 2001) ..... ...... .... .................... .. ..... ........................................ ......................................... ................ 63 Figure 18: Summary of Average household Benefits from CAMFPIRE .. ................... ...... ............ ................. 63 Figure 19: Illustration of Trends in CAMPFIRE Association Expenditure (USD) ...................... .................. 134 Figure 20: Illustration of Costs, Income and Financial Self-Sufficiency of CAMPFIRE Association (ZD) ... 135 Figure 21: Illustration of Costs, Income and Financial Self-Sufficiency of CAMPFIRE Association (USD) 135 NRMPFIN:\I.REPORT 2')JIII.Y200J 3 LIST OF TABLES Table 1: Zimbabwe: Macro-Economic Indicators 25 Table 2: Membership of PET and PMT 31 Table 3: Assessment of Viability and Level of CDF Projects 36 Table 4: Overall Budget for NRMP 43 Table 5: Summary of the Regional Funding Component of NRMP 43 Table 6: NRMP II Budget as Modified After Mid-Term Evaluation 44 Table 7: Breakdown of NRMP II Budget By Project Components and Implementing Agency 44 Table 8: USAID-Zimbabwe NRMP II Variance Analysis 45 Table 9: Breakdown of USAID-Zimbabwe Disbursements by Activity for sub-Contractors 45 Table 10 Summary of Expenditure Administered by Development Associates (15 June 2003) 46 Table 11 ' Overall Summary of NRMP II Expenditure 47 Table 12: Rough Input-Output Assessment of NRMPII 48 Table 13. Natural Resources Monitors and CAMPFIRE Promoters Recruited after 1997 50 Table 14: CAMPFIRE Baseline Data 53 Table 15: Establishment of Wildlife and Other Protected Areas in CAMPFIRE Districts 54 Table 16: Land and/or Resources Protected by Ecotourism and Other Projects 55 Table 17: Trends in Average Trophy Size of Elephants on State and Communal Land 56 Table 18: Summary of High End Tourism Joint Ventures in Communal Areas 60 Table 19: Summary of CAMPFIRE Revenues and Ward and Household Beneficiaries 61 Table 20: Contribution of different Enterprises to CAMPFIRE Revenues (USD) 64 Table 21 : Status of Large CDF Projects as at 30 May 2003 (Excluding ICBs) 65 Table 22: Status of CDF Small Projects as at 30 May 2003 65 Table 23: Status of Eco-tourism Projects 65 Table 24: Responsive of CAMPFIRE Association to Recommendations on Eco-Tourism Market Assessment by Ebony Consultants 66 Table 25: Distribution of CDF Projects 67 Table 26: Status of Community Trusts as at May 2003 71 Table 27: Summary of National Database for Wildlife and Natural Resource Management 86 Table 28: Summary of POMS Results from 2001 Survey 89 Table 29: Summary of Approval Dates of ICB and CDF Projects 94 Table 30: Assessments of the Time Lags in the Processing of Randomly Selected CDF Large Grants 95 Table 31 : Summary of Large CDF Diversification Projects 96 Table 32 : Participation at DNR's CAMPFIRE Initiation Workshop 118 Table 33: Summary of ZimTrust Budget under NRMP II 128 Table 34: Training on PRA philosophy and methodologies 129 Table 35' Key Project Assumptions for NRMP II (1994) 130 Table 36: NRMP II Assumptions and External Factors 131 Table 37: Summary of the Finances of CAMPFIRE Association (1996-2002) 133 Table 38: Categories of Community-based CDF Projects 138 NRMI'FINAI REI'OR"I 2<)JlJl Y2003 4 A. Scope of Work B. Team Composition c. Persons Met List of Annexes D. References and Documents Consulted E. Constitution of the CAMPFIRE Association F. Feedback to Preliminary Presentation of Findings and Recommendations by Team G. Field Trip Reports 1. Muzarabani, Guruve, Chiredzi and Chipinge 2. Gokwe North, Binga and Nyanga H. CAMPFIRE Programme Zimbabwe. Programme Strategy Workshop 1992 I. Financial Summary of CAMPFIRE Association J. CAMPFIRE Baseline data: Summary of Wards, Populations and Areas Benefiting from CAMPFIRE K. CAMPFIRE Offtake Quotas and Values L. Economic Data 1. CAMPFIRE Income by Source 2. Allocation of CAMPFIRE Revenues 3. Allocation of CAMPFIRE Revenues Cumulative by District (1989-2002) 4. Detailed Summary of Ward Revenues 5. Household Dividends by Ward by Year (USD) 6. CAMPFIRE Revenues and Beneficiaries M. NRMP Project Data 1. Summary of USAID-Zim, DA, and Regional Grants 2. USAID NRMP II Disbursements by Function and by Agency N. Macro Economic Data 1. National Populations and Employment 2. Exchange Rates 3. Tourism Data o. List of Projects funded by CDF P. CAMPFIRE Guidelines and Dr Nduku Letter Q . Summary of WWF NRMP II Performance NRMPFINAI.REPORT 29JULY2003 5 AA Action Assoc. ART BRC CA CAMPFIRE CASS CBO CBNRM CCG CDF CITES CSPs DA DEAP DNR DNPWLM FC GOZ IC ICB JICA LT M&E MLG MLGRUD MLGPWNH NACOBTA NGO NR NRM NRMP NORAD PAC PET PMT POMS PRA PWC ROC SADC SAFIRE SASUSG SCI SGI S01 ST TA USA USAID USD USG VIDCO WADCO WISDOM WWF-SARPO ZD ZATSO ZPG&HA ZHA ZimTrust Appropriate Authority Action Magazine Association Abbreviations Africa Resources Trust Bilateral Review Committee CAMPFIRE Association Communal Areas Management Programme for Indigenous Resources The University of Zimbabwe's Centre for Applied Social Sciences Community Based Organisation Community Based Natural Resource Management CAMPFIRE Collaborative Group CAMFPIRE Development Fund Convention for the International trade in Endangered Species CAMPFIRE Service Providers Development Associates District Environmental Action Plan Department of Natural Resources Department of National Parks and Wildlife Management Forestry Commission Government of Zimbabwe Institutional Contractor Institutional Capacity Building Japanese International Development Agency Long term Monitoring and Evaluation Ministry of Local Government Ministry of Local Government, Rural and Urban Development Ministry of Local Government, Public Works and National Housing Namibian Community Based Tourism Association Non Government Organisation Natural Resources Natural Resources Management Natural Resources Management Program Norwegian Agency for International Development Problem Animal Control Project Execution Committee Project Management Team Process Orientated Monitoring System Participatory Rural Appraisal PriceWaterhouseCoopers Rural District Council Southern African Development Community Southem Alliance for Indigenous Resources Southern African Sustainable Use Specialist Group Safari Club International Sustainable Growth Initiative StrategiC Objective One Short Term Technical Advice United States of America United Stated Agency for Intemational Development United States Dollar United States Government Village Development Committee Ward Development Committee WISDOM Foundation (Wildlife in Sustainable Development of Mankind) World Wide Fund for Nature, Southern African Regional Programme Office Zimbabwe Dollar Zimbabwe Association of Tour and Safari Operators Zimbabwe Professional Guides and Hunters Association Zimbabwe Hunters Association The Zimbabwe Trust NRMPFINALREPORT 29JUL Y2003 6 Executive Summary This report presents the findings of the Final Evaluation of the USAID funded NRMP II to Zimbabwe's CAMPFIRE programme. It is important to distinguish between the two. CAMPFIRE IS a long￾term programmatic approach to rural development that uses wildlife and other natural resources as a mechanism for promoting devolved rural institutions and improved governance and livelihoods. Its cornerstone is the devolution of the rights to benefit from, dispose of and manage natural resources. NRMP, by contrast, is a Project with a fixed time horizon. In support of CAMPFIRE, NRMP II has involved two phases. NRMP I spent USD7.6 million and supported four districts in Matabeleland (Binga, Hwange, Tsholotsho and Bulilimamangwe) as well as wildlife management in the Hwange-Matetsi Parks and Wildlife Complex, and some communication, training and research. In 1995, USAID refinanced NRMP II with USD20.5 million, rolling it out as a national project in support of the CAMPFIRE programme. In addition to supporting wildlife management, community development, research and training, NRMP II introduced a grant facility to which communities could apply called the CAMPFIRE Development Fund. CAMPFIRE CAMPFIRE was designed by the Department of National Parks and Wild Life Management in the mid￾1980s, and was initially implemented with limited funding by creating a partnership with three NGOs and the Ministry of Local Government. This grouping - the CAMPFIRE Collaborative Group - then created the CAMPFIRE Association, and in 1991 passed the leadership of the programme to the Association with the intention of building its capacity as the legitimate representation of its rural constituency. As it happened, the CCG members were marginalized in the second phase of NRMP II, with the CAMPFIRE Association's composition of RDCs (rather than communities), technical capacity and financial sustainability becoming questions central to the long-term maintenance of the necessary support to "producer communities". Early Devolution Community Based Natural Resource Management (CBNRM) is now a well recognized, if still controversial, response to the challenge of sustainable conservation. Much of the early impetus to the CBNRM movement was provided by CAMPFIRE, with the key breakthrough being legislation that allowed DNPWLM to devolve "appropriate authority status" to twelve Rural District Councils in 1989 and 1990. Even at this time, DNPWLM recognized that success depended on devolution to sub-district levels, and preferably to the level where communities could meet face-to-face. However, in the absence of legal personality below the level of the ROC, DNPWLM was forced into administrative devolution whereby the legal rights to wildlife were devolved to RDCs, but on the condition that rights and benefits were further devolved to what were termed "producer communities". Good Performance Despite Macro Economic Environment This evaluation took place at a time when Zimbabwe was undergoing severe political and economic hardships, with a shrinking economy, controlled priCing and a highly distorted market in foreign exchange. The most important finding, therefore, was that despite these difficulties, CAMPFIRE was still working remarkably well at the level of the ROC and particularly the community. This is a testimony to the efficacy of the principles upon which the programme is based, which appear sound and robust even in trying circumstances, and to the considerable effort that has gone in to creating community capacity, and especially an acceptance of the efficacy and legitimacy of local level management and benefit. USAID funding was important, particularly in expanding the concept of CAMPFIRE beyond the original dozen wildlife districts. Positive Conservation Impact There is ample evidence of sound wildlife and natural resource management. In the face of rapid population growth that has over-run large areas on the Sebungwe and Zambezi Valley, the CAMPFIRE programme has protected an area of wild land roughly equivalent in extent to the Parks and Wildlife Estate (i.e. some 50,000km2). In the ten years since its inception, wildlife populations appear to have increased by about 50%, with elephant doubling for 4,000 to 8,000 in CAMPFIRE Areas. The primary impetus for conservation has been safari hunting. Robust systems of management have been developed and widely adopted at community level, with quota-setting being especially important. Field visits confirmed that most community leaders understand and are able to set and manage quotas and safari NRMPFINA LREPORT 29JUL Y2003 7 hunting. In at least seven districts, electric fences and other measures are being to reduce the of problem animals, at least four districts have fire management programmes, and In numerous there is evidence of more care being taken to manage and control natural resources. In Mazowe, fishing by-laws are being enforced, and the natural vegetation has been in the middle of Mahenye community. Much of the credit for this must go to WWF, and Its systematic approach to working with communities to develop appropriate wildlife and natural resource management technology, and then rolling this out with manuals, trainer's manuals, and a carefully designed training methodology that includes site-level training but places most emphasis on the training of trainers. Approximately 18% of NRMP II funds have been spent on improving natural resource management, and the conclusion is that the impact of this money has been high, largely because it has been channeled through people who are recognized experts in their field. Economic Impact The economic impact of CAMPFIRE has also been positive. Methods for involving communities in the selection of their private partners using open competitive marketing were developed in the early 1990s. Initially, these rapidly increased the income from wildlife concessions and latterly, in the face of pressures to the contrary, they have maintained these prices and have largely excluded corruption. Together with the increasing wildlife populations mentioned above, revenues from safari hunting increased rapidly to USD2 million annually. Since 1989, CAMPFIRE has generated direct in come of over USD 20 million, with an economic impact of USD 100 million. Diversification At least twelve high-end tourism lodges have also been developed in communal areas but the revenues from these, and also from secondary activities such as employment, tourism multipliers and the provision of natural products are generally not captured in the CAMPFIRE revenue database. Part of the reasons is that some RDCs treat this income as General Revenue. A study commissioned by CAMPFIRE Association (PriceWaterhouseCoopers, 2001), suggests that natural resources such as timber, fishing and sale of products including sand actually provide some 30% more money than is reflected in the CAMPFIRE accounts. Unfortunately, much of this does not get returned to communities. Especially after the 1998 Mid Term Review, NRMP II attempted to diversify the commercial base of CAMPFIRE. Some 10 ecotourism ventures were established as were 4 for managing natural products such as bamboo, fishing and mopane caterpillars. These initiatives got off the ground slowly, and their timing coincided with Zimbabwe's economic collapse. While a good proportion (5 and 3 respectively) are likely to be sustainable, at least if the economy returns to normal, they would have benefited from more and better technical evaluation and support, earlier encouragement of private sector partners, better background assessments of the viability and value-added of such projects, and more consideration of the need for coordinated support such as product development and marketing especially as effective diversification of NRM is difficult. Project design may have under-estimated the difficulties of diversification, and was hampered by the declining capacity of the CCG and the declining tourism economy. USAID's procedural requirements were also too cumbersome for community implementation. Implementation was also hampered by working largely with CA and RDCs, whereas working directly with communities may well have been more effective, as was demonstrated late in the project including by the small grant window. This resulted in a situation towards the close of the Project where tough measures were required to complete many projects before the funding window closed, and this is not always compatible with the process of community development. Nevertheless, we endorse the efforts in this direction, and note improvements such as the development of a small grant window (albeit five years into the programme), a much higher level of tourism and engineering expertise, more insistence on community-private partnerships, and a greater focus on communities Including the development of some 16 trusts to manage these projects. An over-sight in this programme was the importance of product development and marketing. While data suggests that CAMPFIRE is still 95% dependent on safari hunting, this underestimates the level of diversification by approximately 30%. Safari hunting has proven to be robust both environmentally and economically, even in extremely difficult circumstance. It is the reason that many CAMPFIRE communities can be considered sustainable, especially in the original twelve wildlife districts. Diversification, while desirable, is a process that requires a long gestation period and considerable investment In product development. In the short term NRMP's investment was critical to CAMPFIRE's expansion and acceptance NRMI'FINALREI'ORT 29JUL Y2003 8 at a time when the programme was under political pressure. It enabled the spread of the CAMPIFRE principles into non-wildlife areas, bringing with it wider political support of the CAMPFIRE process, as well as economic and conservation benefits. Fiscal Devolution CAMPFIRE is build around the concepts of fiscal devolution and local responsibility. From 1989 to 1995, the amount and proportion of revenues devolved to producer communities increased rapidly, and from all reports this was the primary impetus for both the wildlife conservation discussed above and for rapid improvements in community institutions and governance. However, from 1996, the rate of devolution plateaued and from 2000 the process began to reverse itself. By 2001, only 38% of revenue was returned to producer communities, some 20% was used for CAMPFIRE management, but over 40% was retained by ROCs for general purposes compared to an agreed upper limit of 15%. A few districts have bowed to the pressure of the majority of councillors, and are spreading CAMPFIRE revenues thinly to a large number of people. This risks earlier successes. We quote the example of Sidinda Wards in Hwange. The community immediately stopped poaching once they received benefits, and the populations of buffalo and sable increased steadily. However, with only one vote in Council, benefits were then shared out amongst eight Wards. The Sidinda community no longer has a significant conservation incentive and poaching has resumed. While we quote these problems, the concept and level of devolution in many districts is still strong, and we confirm the strength of the correlation reported by ZimTrust (2001) between fiscal devolution and institutional development. We also note that an important benefit of the spread of CAMPFIRE through NRMP II COF investment is that these prinCiples are being adopted in almost all projects, and what was once a contentious principle is now an accepted norm. Innovation is also occurring. In Chiredzi, for instance, the hunter pays 60% of hunting revenues directly to the village in which an animal is shot within seven days of the hunt. ROC technical staff in several districts recognize the power of such innovations, but are not always supported by councilors. Chipinge ROC, for example, agreed that after a decade of supervising the Mahenye community, it was now time to give them full authority. Rural District Councils, Services and Taxation In general, the competence and level of commitment by staff in ROC CAMPFIRE Units was strong, and technical and equipment support provided by NRMP II had created worthwhile capacity. Wildlife districts are likely to be able to maintain much of this capacity using 20% of revenues, although we believe that it would be beneficial to the programme if staff such as resource monitors were employed at community level and this proportion was reduced to 15%. We also recommend that revenues are paid directly to communities, who then pay the ROC on the basis of services. We are concerned at the high level of taxation of producer communities. This includes the official 15% levy by ROCs (plus an additional 25% retained informally) and the 4% levy for CAMPFIRE Association. We strongly recommend that the total figure is kept below 10% and is in line with the taxation of other resources such as cotton or livestock to avoid differential taxation of wildlife. Intuitional Development One of the most remarkable achievements of CAMPFIRE has been the strength of institutional development at the level of producer communities. Every year these communities receive their wildlife dividend, and all indications are that use is decided democratically, that people retain and sometimes use their right to have household cash benefits, and that a large number of projects are implemented sensibly and well. Finances appear to be reasonably well managed, with transparency and peer review to a large extent preventing misuse. The system is undoubtedly working, and is an inspiration to rural development. Lost Opportunities in Institutional Development Particularly at producer community level, CAMPFIRE provides an important and rare test case of the emerging interest in fiscal devolution 1 and rural governance. Our disappointment is that more has not been made of the opportunity to monitor and develop this experience, to use it to innovate appropriate technology in the manner that has been done for natural resource management. CAMPFIRE provides a cutting edge example of the powerful impact of fiscal devolution, and had it been developed and spread in the manner of devolved wildlife management systems (e.g. quota setting, community-private contracting, revenue distribution, etc.), the benefits to rural development in general are likely to have been huge. I For example, the World Bank (undated, 2001?) emphasises this, whereas Ribot (1999) in a review of such programmes in Africa over the past 40 years is unable to find examples of fiscal devolution. NRMPFINALREPORT 29JUL Y2003 9 Perhaps the greatest favour CAMPFIRE can do in the cause of devolved rural governance is to monitor, adaptively develop and record this experience. Our field assessment, inclu?ing interviews, inspection. of projects and of financial records, convinced us of the efficacy and power of this approach. However, unlike wildlife management and economic benefits, we have been unable to support our arguments with data. As with wildlife management, the key to developing institutional management systems will lie in the use of people who are recognized experts in this field. While CAMPFIRE, and especially ZimTrust excelled in advocacy and awareness creation, the programme always lacked the requisite professional expertise in institutional development. Regional and International Impact CAMPFIRE has also had a considerable regional and international impact, and we consider the returns on the investment made by NRMP II in this aspect (approximately 10% of project expenditure) to be very high. CBRNM is now generally accepted and practiced in southern Africa, and CAMPFIRE has contributed significantly to this experience, exporting technology and lessons. At an international level, CAMPFIRE has been influential in supporting the rights of local people to benefit from and manage their own natural resources, and in developing an understanding of CBNRM and common property management. Centralization and Excessive Taxation We turn now to some of the more negative aspects of NRMP II. We have already noted the increasing propensity of ROCs to extract revenues from communities. It is a pity that NRMP worked through and strengthened the centralizing hand of ROCs (although we have noted how some ROCs, and especially their CAMPFIRE units, have tried, in extremely difficult circumstances, to devolve revenues). We accept that project design was heavily influenced by the general political centralization of Zimbabwe, which exaggerated the prominent role given to ROCs and CA in implementation. However, project effectiveness and impact on the ultimate beneficiaries would have been considerably enhanced by working directly with producer communities rather than through these intermediaries. CAMPFIRE Association The project has put considerable efforts into strengthening CAMPFIRE Association as the lead agency. This has been less productive than anticipated. CAMPFIRE Association never evolved real leadership capacity nor, as had been hoped, into an association truly representative of rural producer communities. Its membership still consists entirely of ROCs, and remains too similar to the Rural Oistrict Councils Association of Zimbabwe. CAMPFIRE Association's expectations of taking on many implementation and monitoring functions were always over-optimistic, and this was exacerbated by the unexpectedly slow improvements in CAMPFIRE Associations' management capacity. CA also deliberately marginalized NGOs from the programme. Our recommendations are that CA needs to diversify its membership to included producer communities, private producer communities (e.g. Conservancies) and other supporters (e.g. NGOs). It also needs to diversify and strengthen Board membership and experience (e.g. lawyers, financial managers and industry leaders). This would improve its financial sustainability, technical capacity and strength of leadership. We also recommend that CAMPFIRE Association carefully redefines its mandate and membership to increase accountability and services to producer communities. CAMPFIRE Development Fund Grants A significant amount of NRMPII finances have been used to develop the capacity of ROCs to support CAMPFIRE activities and to implement COF-funded projects, largely through support to an Institutional Contractor and the CAMPFIRE Association. On the one hand, this has enabled CAMPFIRE to spread country wide, with Institutional Capacity Building Grants (including training, vehicles and equipment) to 23 districts and an increase in membership of the CAMPFIRE Association to 52 of Zimbabwe's 57 ROCs. This has been crucial for the political survival of CAMPFIRE in what was perceived as a white-dominated technical initiative, and is therefore a worthwhile investment almost regardless of cost because it lays the social foundation for the programme. conceptually, technically and financially this has been the weakest aspect of NRMP II. The use of is, in case, difficult, but these difficulties were exacerbated by the marglnalisatlon of the NGOs, their loss of capacity independently of NRMP, and other problems. This maintained the COF as a fairly ordinary grant-making mechanism where, in the correct environment, CAMPFIRE provides a foundation for considerable innovation and experimentation. In the circumstances, NRMPFINALREPORT 29JUL Y2003 10 the efforts to implement the CDF have been steady, albeit with cumbersome mechanism related to USAID procedures. The capacity developed in RDCs by the USD2 million ICB grants is likely to be maintained, at least in wildlife districts, and the USD1.3 million CDF grant has developed over twenty diversification projects, many of which are sustainable at least I normal economic circumstances. Had circumstances been normal and more conducive to devolution and the use of the private sector, we would have criticized the CDF for not emphasizing loans. performance-based grants, private-community partnerships, the development of sinking funds or an endowment mechanism and, above all, a direct focus on working with beneficiary communities. Indeed, the latter stages of the CDF, when the small grant window was opened and projects were targeted directly at communities, yielded superior results, which does pose the question of why this was tried so late into the programme. Generally, results have been mediocre. Delivering financing directly to communities remains the right objective, and this experience should be built on. It is a pity that funding is being withdrawn, just as sufficient experience is emerging on how to improve delivery. Loss of Intellectual Leadership During NRMP, the CAMPFIRE Collaborative Group has dissipated with the loss of innovation and leadership within the programme. To some extent this has been external to NRMP. However, it may also be consequence of the heavy funding of CAMPFIRE Association, which believed it could go it alone and seldom actively supported NGO-partners in their quest for funding. Other Project-driven factors were the increasing domination of the CDF in its latter stages, and also the reliance on short-term consultants, which, to some extent replaced critical long-term relationships between facilitators and communities. In several communities visited, this was seen as the biggest threat to sustainability. However, the NGOs concerned must also shoulder a good deal of the responsibility, with all of them independently losing technical capacity. One criticism leveled at some (not all) of the NGOs is that they have been provided with a lot of equipment by USAID, and access to funding, yet provided little support to CAMPFIRE after their core funding grants ran out. ACTION Magazine for example, quoted CAMPFIRE Association full market rates although their equipment had been grant funded through NRMP. Project Delivery Challenges Heavy funding changed the nature of the programme, especially in the last three years. It empowered CA relative to the partners. It also increased the emphasis on CDF projects - in all five regional reports presented at the 2003 CAMPFIRE Annual General Meeting the only topic reported was the status of CDF projects. In this period, CCG ceased to provide effective programme leadership so, de facto, the key grouping in the programme became the PMT. The PMT is a technical body to review CDF projects, and not well suited to this role. In preparation for NRMP II, PET had been designed with a careful balance between ROC, government and civic actors to control the flow of resources from NRMP II to communities. However, in interpreting leadership as domination of the programme, the CAMPFIRE Association discouraged PET. NGO participation was weak because of this, but also because of their own increasing capacity constraints. PET did not work, for reasons of personality rather than deSign, and was therefore replaced by the PMT as the mechanism for administering and distributing projects, which also centered more power with CA. This worked reasonably well as a technical mechanism for project implementation. However, the CAMPFIRE programme lost cohesive, cooperative leadership and vision. Membership of the PMT comprised only government or quasi-government agencies (i.e. RDCs, DNPWLM, MLGPWNH, and USAID/DA) and lacked civic and technical peer review. Moreover, it was comprised predominantly of ROC persons, yet was responsible for sharing out projects amongst this membership. An unsurprisingly the result was that each ROC was entitled to one large project. The majority of PMT was more concerned with handing out projects as quickly and with as little difficult as possible. ThIS placed both USAID and the Institutional Contractor in the difficult position of having to veto unviable projects, which in the c,ase of the latter conflicted seriously with its second role of providing capacity to the CAMPFIRE Association. There was also insufficient expertise or business experience in the PMT. While accepting that the CDF was designed in an antagonistic environment, noting that it was important for expanding the CAMPFIRE programme and diversifying its economic base, and noting than the CDF was a useful (if not innovative) contribution, we are nevertheless critical of several design features. First, it provided too much free money. Therefore projects were allocated one to each district (each member of the PMT got one project), equity criteria dominated technical or economic factors (although DAIUSAID NRMPFINALREPORT 29JUL Y2003 11 constantly emphasized the projects must be sustainable), and the process lacked the discipline and incentivisation of either loan repayment or performance criteria. Second, we question the process whereby RDCs were given Institutional Capacity Building Grants (USD2 million), and only three years later CDF Grants (USD1.3 million). In contrast to this typical blue-print approach, we suggest adaptive management using pilot projects and a learning process may have reaped quicker results. We note the relative inefficiency of the earlier period when projects were channeled through RDCs compared to the improvements introduced by the small grant window and direct funding to community level. However, we also wonder why it took five years to initiate the relatively simple small grant window. With the funding of NRM projects being universally problematic, an interesting opportunity might have been to design it as a learning opportunity as well as to deliver diversification. However, we also note that the loss of capacity and synergy within the CCG, weakened the CAMPFIRE programme's ability to create and use learning processes. Thirdly, and linked to this, we are concerned that the stiff criteria imposed by USAID limited the appropriateness and flexibility of the CDF. The simple and community-friendly guidelines developed in the first year of the Project did not fulfill USAID's administrative requirements, and it was four years into the project before the CDF Guidelines were finalized. These guidelines were cumbersome, and are an example of the bureaucratic tail wagging the developmental dog. Fourthly, the short implementation period (at least in the time scale of rural development and institutional change), and the fact that this funding mechanism was not extended through a sinking or endowment fund, was not ideal. Opportunities to build on emerging lessons could not be used. The funding profile did not matc the rate of uptake by communities. And the funding profile did not match the requirements of diversification or product development, which is also a long-term process. The grant approval process was slow and cumbersome, a large part had to do with meeting USAID financial review requirements before a grantee could receive project funds. It also tackled a difficult areas (ecotourism and NRM diversification). Consequently, delivery was slow. Almost half of the CDF projects are still incomplete although USAID and DA estimate that 80% will be completed. Initially intended to deliver USD6 million to communities, only USD 1.3 million was actually spent on CDF projects (6% of NRMP II funds), with a further USD 2 million spent reasonably effectively to support RDC CAMPFIRE capacity. In this regard we must state that the management capacity of communities is not the limiting factor. Every year they invest over USD1 million in a large number of highly participatory community projects, suggesting that over the project lifespan they have invested seven times as effectively as the project in terms of the amount of money spent, and probably considerably more than this in terms of value added. The problem lies in the structure and complexity of the CDF, and the pOlitical circumstances that exerted pressures to use it as a centralized, grant facility. For these reasons, it could not be adapted to effectively reach and use sub-district capacity, and initiating this towards the end of the project required a lot of hard work in difficult circumstances. We have repeatedly stressed the importance of the CDF in ensuring the spread of CAMPFIRE and its political survival. This alone is enough to justify this funding. While, generally, we might not support grant funding allocated in this manner on developmental grounds, in the difficult circumstances of the time our conclusions is that the CDF had far greater positive effects than drawbacks. We nevertheless suggests that, as a financing mechanism, the CDF was steady but unimaginative and that, for whatever reasons, opportunities to create further innovation around the CAMPFIRE programme were not taken. The need for long lead times, and strong technical support to a learning process, are an important lesson of the Project. Future Funding Mechanisms Should be Goal Specific We have suggested that any future funding should be structured much more carefully. If the intention is to promote devolution, funding should be matched directly to the amount of money devolved to community level. If the intention is to promote enterprise diversification, funding should support private-community partnerships in the form of loans (which also allows the possibility of a sustainable fund) and should introduce the formal banking system using an under-writing systems. If the intention is to support social or environmental projects, grants should use stringent criteria for involvement and benefit, and should also reward performance (as is now done commonly with social investment funds where). NRMPFINALREPORT 29JUL Y200J 12 Conclusions In conclusion, NRMP II and CAMPFIRE confirm the concept that the devolution of natural resource management can be a highly effective mechanism for the participatory management of wildlife and natural resources, and for improving local institutions and governance. The time frame of donor projects is usually too short for the behavioral and institutional changes that are central to this process, and heavy levels of funding have to be managed extremely carefully to avoid upsetting these delicate and important processes. The timing of the withdrawal of NRMP II is unfortunate, given the importance of supporting CAMPFIRE during a period of serious economic decline, and the opportunities to learn from what has happened. Nevertheless, while economic circumstances will set CAMPFIRE back, it appears robust enough to survive. NRMP II funding has been 'extremely effective where it has capitalized on the presence of long-term recognized expertise to develop natural resource management systems (18% of budget) and to research and promote the concept of devolved community management (10%). Institutional capacity building (25%) has been weaker, largely owing to the absence of such expertise, and represents a significant lost opportunity. While management (40%) and funding of the CDF (7%) has spread the CAMPFIRE concept broadly, better design and monitoring on this component would have significantly increased the level of benefit and devolution. It is important to place these comments within the institutional dynamics of the project and to note that (in agreement with DA) over-ambitious project design, institutional rivalries, reluctance to use to technical assistance and the private sector and to innovate, and uneven quality of TA contracted by some Project participants, were also important constraints. In order to promote the original CAMPFIRE principles of devolution of authority over natural production systems to producer communities, CAMPFIRE stakeholders need to emulate regional experience to press for legislation to provide for appropriate authority status, and even land rights, at sub-district units of decision-making, and preferably at the level of the village. Effort to promote local-level proprietorship, such at the experimentation with community trusts, must be continued. Community and district capacity is not a limiting factor, but to achieve devolution the long term leadership of the programme needs to be strengthened. The greatest contribution of the CAMPFIRE programme has been the lesson that fiscal devolution leads to improved rural democratization, governance and management. Building improved organizational technology around these opportunities, and monitoring and documenting these lessons, might go a long way to demonstrating the value of devolution. To complete the executive summary, we summarise the positive impacts and weaknesses of NRMP. NRMPFINALREPORT 29JULY2003 13 Summary of Positive Impact 1. Conservation impact In the face of 10% population growth in some areas, and doubling of rural populations between 1980 and 2000: 1. Wildlife areas were demarcated (often informally, sometimes with fences) and were largely maintained (but area/size data not available) . 2. Elephant populations increased steadily, and buffalo populations were maintained (despite crash in 1992 drought) 3. Trophy quality was largely maintained 4. With diversification, small "community conserved areas" were established or protected in non￾wildlife ecoregions - Gairezi, Vhimba, Mavuradonna, Mazowe Mountain, caves in Matobo, etc. 5. Reduced or contained veld fires in Chiredzi, Chipinge and Gokwe North 6. Poaching contained or reduced levels of illegal off-take (wildlife populations, but data on fish not available) 7. Anecdotal narrations suggest reduction in tree-felling (Nyanga, Ntanjambili, Maitengwe) 2. Economic benefit Impact 1. Revenues from hunting were increased rapidly and then maintained at USD1.5-2.0 million annually although NRMP II moved away from promoting hunting due to policy reasons; 2. At least 12 high-end ecotourism projects are operational in communal areas - Ngami, Gorges, Bumi, Mahenye, Binga (but no income data) 3. Increased number of households (from 7,861 in 1989 to over 80,000 in 1993) benefited directly (in cash) from wildlife and many of these communities made social investments and built small businesses (no full records available) prior to NRMP and continued throughout the NRMP period. 4. Secondary benefits in terms of services, goods and supplies, employment creation at local levels around successful tourism projects can be assumed (no data available) 3. Diversification 1. At least 18 middle range eco-tourism ventures were started (of which 4 have been completed - Nyanga, Vhimba, Banje, Gonono) 2. New areas of revenue were promoted (e.g. fisheries in Beitbridge, Mazowe; honey collection in Nyanga, Gokwe South, and Masau (fruit) in Muzarabani and Guruve) 4. Devolution 1. CAMPFIRE movement and its supported structures have led to increased awareness of entitlements and rights and demand for these at local level (Chitsike Report mentions demand and the team's impression from site visits confirms this) 2. At least 17 trusts established at sub-district level and 7 are functioning (bank accounts, regular minuted meetings, paid employees) 3. In 1 hunting District (Chiredzi) and 2 Eco-tourism projects (Gairezi and Karunga Trusts) payments are being made directly to sub-district level 4. In 5 out of 7 sites visited by evaluation team, Council and CAMPFIRE unit officials strongly supported devolution and were taking measures to implement it (but reportedly there is less commitment by Councillors with some notable exceptions) 5. 2002 CAMPFIRE Revenue Guidelines reinforce principles of administrative devolution 6. In 6 out of 7 RDCs visited, there was full acceptance of the right of communities to utilize income as they saw fit. 5. Institution and Capacity Building at Sub-District, District and National Levels 1. Over 100 village and ward CAMPFIRE committees in 23 Districts learned basic organizational skills - book-keeping, minutes, meetings, etc. 2. In at least 13 districts, natural resource monitors and community leaders in particular learned wildlife management skills - setting quotas, selling wildlife, monitoring hunting, managing electric fences, problem animal monitoring, counting wildlife, ecological management 3. Systems of controlling off-take of natural resources were put in place - e.g. fish poachers were fined, with fines paying guards and providing community benefit 4. Fire management implemented in 4 areas (Chipinge, Chiredzi, Gokwe North and Guruve) 5. CAMPFIRE Support Units established at ROC level in 23 districts and are still providing training to SUb-district levels in Wildlife districts NRMPFINALREPORT 29JULY2003 14 6. Over 100 village and ward wildlife committees established and functioning 7. NRM by-laws and constitutions developed for use in the village and ward wildlife committees 8. Training of members of Board of Trustees 9. Asset management and books of accounts kept in order at RDC units 10. In most producer wards basic record keeping is in place 11. Communities undertaking project identification and implementation in wildlife districts (but weak monitoring and no centralized database) 12. CAMPFIRE Units in wildlife districts generally operate well, maintaining expenses at 20% (but could devolve more functions to communities and reduce this percentage to between 10-15%). 6. Governance 1. High level of transparency and community participation in revenue distribution and use at village and ward levels 2. CAMPFIRE structures have provided a forum for high levels of community participation in decision￾making on NRM and other issues 3. Generally, funds were well accounted for at sub-district level and in RDC CAMPFIRE Units 4. CAMPFIRE committees at village and ward levels are elected 5. Transparent flow of information at ward and village and between some wards and villages concerning issues, action plans and projects 7. Technologies and Innovation 1. Development of cutting edge knowledge about the process of empowerment (e.g. quota setting, contract negotiation with private sector etc.) 8. National, Regional and International Impact 1. With broadening of base, CAMPFIRE enjoyed a growing social and political acceptance at national level (by Parliamentarians and within Central Government) 2. NRMP II ensured survival of the earlier CAMPFIRE prinCiple (see Constitution of CAMPFIRE Association and the CAMPFIRE Principles) to a large extent by funding its expansion to non-wildlife districts in the country 3. Consumptive forms of wildlife utilization and non-consumptive use of wildlife (eco-tourism) on communal lands became acceptable forms of land use 4. CAMPFIRE principles have proved resilient and have spread to non-wildlife areas 5. Other countries in the SADC Region (Botswana, Namibia, Zambia, Mozambique) have adopted CAMPFIRE principles adapting methodology and approach by devolving authority to sub-district community organizations rather than district level administrative units 6. Technologies developed by CAMPFIRE were widely adopted in the region 7. CAMPFIRE contributed significantly to promote acceptance of sustainable use principles at the international level 8. CAMPFIRE put Zimbabwe squarely on the international scene by its activities in CITES (particularly 1997), becoming a fertile ground for research and learning and producing a vast amount of literature (over 300 publications) on CBNRM and common property resource management. 9. Sustainability 1. Systems, infrastructure and capacity are being maintained in wards and villages receiving revenue from wildlife utilization, and in some community eco-tourism initiatives 2. CAMPFIRE units are probably sustainable in wildlife districts 3. Most CSPs have diversified funding base and several have made a shift towards regional activities 4. At this point in time the team has concerns over the long-term sustainability of some of the CDF￾funded activities, particularly those ecotourism projects with weak market links and those projects that have not yet been completed. NRMPFINALREPORT 29JULY2003 15 Summary of Weaknesses of NRMP II The following weaknesses have been noted by the evaluation team: 1. RDC Performance 1. NRMP has strengthened RDC control and role as gatekeepers in relation to producer communities 2. Poor management by RDCs of approved community projects 3. Misuse of project vehicles in some RDCs (Muzarabani) without monitoring and corrective mechanisms 4. In the face of the current economic crisis, RDCs retaining an increasing amount of revenue, thereby disadvantaging the intended primary beneficiaries of the NRMP Project and the CAMPFIRE Programme 2. CAMPFIRE Association Performance 1. Current membership of CA does not include producer communities and conservancies 2. Current membership of CA has made no room for associate membership for supporters, promoters and well-wishers of CAMPFIRE 3. Marginalization of collaborative partners by CA from core decision-making forum - the exact opposite to the recommendations of the Mid-Term Review 4. Policy of marginalization of CCG-NGO members and conversion into service providers has removed the opportunities for long term facilitation and partnership with producer communities 5. At times weak leadership of CA Board and Secretariat (strategic plans remained unimplemented, sustain ability issues neglected, poor coordination) 6. No synergies created with private sector 7. The SGI continues in the same flawed vein in proposing the establishment of a CAMPFIRE Institute providing services as an implementer without seriously looking at its income projections, costs and role change implications (no documentation of financial analyses) 8. Discontinuation of team review processes (annual reviews) diluted the powerful dynamic and unity of the CAMPFIRE movement and inhibited NRMP implementation 9. Heavy funding by USAID inhibited diversification of donor base 10. Board membership lacks diversity, specialisation and external input (particularly private sector) 11 . CA has failed to take up policy issues in a formalised manner with Government (devolution of AA to sub-District level, weak follow up on required legislative harmonization) 12. CA should have remained a membership and advocacy agency 3. Devolution 1. No enabling Legislation has been passed to provide proprietorship at producer community level (Traditional Leaders Act provides assembly and boundaries but no entitlement and legal personality) 4. Diversification 1. Compared with total investment inflow, the output (increase in revenue generation from non￾wildlife sources) has been minimal 2. In spite of diversification efforts under NRMP, consumptive wildlife revenue provides more than 70% of the benefits 3. Insufficient investment in product development and marketing 4. Eco-tourism projects were embarked upon with minimal marketing and product feasibility except in cases where collaboration with private sector was already in existence (e.g. Nyanga Downs Fly Fishing Club where the project builds on existing demand and use of the Gairezi River for fly fishing) 5. Project Design 1. Technical criteria and requirements of USAID in CDF implementation far too complicated for producer community projects 2. Insufficient attention and resources were devoted to diversification 3. The thrust to turn a fledgling association like CA to take on implementation responsibilities was questionable given the CA's primary role as an organisation representing the interests of its members and carrying out advocacy on their behalf 4. It was a design flaw of NRMP to expect an RDC membership organization like CA to push for devolution below the district level NRMPFINALREPORT 29JULY2003 16 5. Ratio of direct deliverables to community versus other components was skewed (1 :10) 6. Phasing out of CCG input after 1998 mid-term evaluation threatens long term collaborative partnerships and capacity within CAMPFIRE 7. PMT after 1998 represented only CA, Government and the donor without civic society and private sector 8. CDF did not succeed in rewarding devolutionary process and good performance, although the CDF guidelines had adopted such an approach. Funding was awarded based on each district receiving some support. 6. Project Delivery and Coordination 1. Absence of technical peer review of proposal approval processes (while some CCG members did not want to be part of the process due to potential conflicts of interest, the PMT did not succeed in co-opting external reviewers with appropriate expertise) 2. Training in the Institutional Capacity Building phase 1995-1997 of RDCs/Communities on USAID requirements of accessing CDF funds was insufficient; CDF manual came out late (early 1998) and insufficient orientation and follow up training was done on applying the CDF manual 3. Distribution of projects across districts was not based on sound economic and business criteria but was most often based on equity 4. Role conflict, lack of team spirit and commonality of purpose between key actors in NRMP implementation (CA and DA) 5. DA and USAID were pushed more and more into an unintended implementational role on CA institution building and overall guidance of the CDF by default (CA's own leadership weakness after 1998) 6. CAJDA took 5 years to innovate a small grant window enabling speedier processing of projects 7. No piloting of CDF and accompanying development of sound methodology based on piloting results 8. 7. Project MIS and Monitoring of Socio-Economic Impact 1. No centralized MIS on NRMP implementation and key monitoring aspects of the program are missing or weak (e.g. income data at HH level, socio-economic data, capacity and institution￾building data etc.) 2. No collated and analyzed information to substantiate capacity building effort and income generation at ROC and sub-district levels 3. Despite overall aims· of promoting diversification, devolution, ICB and accountable decision￾making, impact and performance indicators were not developed and monitored either by the Project or the CAMPFIRE Programme 8. Project Sustainabllity 1 . No sustainable mechanism put in place to maintain the database already set up by service providers (e.g. WWF, ZIMTRUST) 2. Unsustainable installation of units at ROC level in non-wildlife districts thereby increasing overheads of the districts 3. Early efforts from 1997 at establishing financial sustain ability were not followed up by CA and many opportunities thereafter missed (creation of endowment fund/trust for CA or NRM at large). 9. Project Collaborating Partners/Service Providers 1. DNPWLM took a back seat or remained passive in program implementation (involvement mainly confined to the PMT), particularly when it came to policy issues, monitoring of wildlife and compliance with conditions of AA. 2. Some of the CSPs did not perform up to expectations (SAFIRE on providing training on trusts, ZIMTRUST on institution building, accessing CDF, application of POMS and financial management, WWF on Decision Support System models, CASS on institutional training, time lags in submitting reports and weak dissemination at ROC level) 10. Role of USAID 1. USAID remained too optimistic about the capabilities of CA in spite of early warnings by DA 2. USAID misread signals of absorption and institutional efficiency at ROC levels NRMl'rINALREI'ORT 29JUL Y2003 17 I. Introduction A. PURPOSE OF THE EVALUATION The overall purpose of the evaluation is to assess the extent to which the Natural Resources Management Program's (NRMP) original goal, purpose and outputs, and any subsequent revisions of these, were achieved over the life of the Program (1995-2003), and to recommend actions that should be considered by stakeholders responsible for the future development of the Communal Areas Management Program for Indigenous Resources (CAMPFIRE). The evaluation assesses the relationship between impact, structure and financing over the entire period of the program but, given the work done in earlier reports (e.g., Mid Term Evaluation), focuses on the post-1998 period in terms of data collection and analysis. This evaluation represents the closing analysis of a thirteen-year investment, and presents an overview of what has happened, problems faced, and what can be done in the future. CAMPFIRE is a world-renowned program. It was an important example in support of the global shift towards what is now commonly referred to as community-based conservation or conservation-based development. The devolutionary principles2 upon which CAMPFIRE is based are well recognised and, where properly implemented, powerful. However, the process of institutional, organisational and political change leading to the implementation of these principles is fraught with challenges. The CAMPFIRE programme is no exception. At the start, we must emphasise that the NRMP is not the same as CAMPFIRE, but is a project designed to build on a highly promising start to roll out the initial intent of the CAMPFIRE programme. While project design and implementation is important, any analysis of the efficacy of NRMP must recognise that its nature has been affected by an increasingly centralising political system, by pressure emanating in the USA to move away from the initial focus on wildlife management and hunting and, latterly, by serious macro-economic problems. This has affected the weighting of the programme in terms of its emphasiS on commercial, biological and political sustainability; wildlife versus diversification; and the ability to translate CAMPFIRE's devolutionary principles into reality. The central objective of this evaluation is to use the CAMPFIRE and NRMP experience to elucidate the links between projects and the socio-political systems in which they are implemented, and especially what worked, what did not, and why. The evaluation uses the key reference documents that provide the framework for the program management, especially the 1998 Mid-term Evaluation and the project documents (NRMP II and 1998 Amendment). It uses the large amount of written, published and unpublished materials associated with the programme. It is supplemented by interviews with key informants and by invaluable site visits to programmes and projects in seven Rural District Councils (RDCs). Finally, all members of the team have considerable experience of CAMPFIRE dating back to its inception, experiences with similar regional and national initiatives, and considerable insight into the reality of the issues involved. To avoid repeating the caveat that NRMP II coincided with a particularly difficult period on Zimbabwe's history from the perspective of devolved natural resource management, we quote an import observation by DA: "One general observation is in order. The Evaluation describes/analyses well the most important aspects of the Project, but does not give a feel for the project dynamics and complexity in terms of the high level of coordination/rationality expected of implementers and other stakeholders matched against institutional rivalries, diversity of purpose, very different levels of organizational development among agencies, sometimes hostile Government oversight/threats, etc. Many decisions made in the course of the project were made on very pragmatic grounds - what can be done rather than what should be done, the least of two evils, etc. The report sometimes reflects this, but seems to conclude that things would have been much better had more tech expertise been used or had better systems been established when they were obviously needed, etc. Project dynamics made it very difficult for this to happen - for example forces were unleashed that resisted TA or otherwise rendered it ineffective, particularly with respect to CA. Institutional rivalries often precluded collaboration. This Project was much more ambitious and 2 Murphree (2000a: 5.) distinguishes between decentralisation and devolution as follows: "Decentralisation is the delegation of responsibility and limited authority to subordinate or dispersed units of hierarchical jurisdiction, which have a primary accountability upward to their superiors in the hierarchy. Devolution involves the creation of relatively autonomous realms of authority, responsibility and entitlement, with a primary accountability to their own constituencies." NRMPFINALREPORT 29JUL Y2003 18 complicated than most donor efforts at the onset and also had to deal with some difficult economic/political externalities during its implementation." Circumstances were undoubtedly difficult. GOZ was increasingly wary of NGOs. CA responded to this, and to the opportunities provided by significant funding, to sideline many of the NGOs who, in turn, demonstrated only weak commitment to CAMPFIRE (with the exception of WWF) and also lost capacity. This was an antagonistic period. At a time it faced strong political challenges, the CAMPFIRE movement lost the leadership capacity to deal with these. The innovating potential of NRMP was seriously affected by these events. It did well to survive, and to ensure the survival of CAMPFIRE. The difficult question to answer is whether USD 20.5 million could have been used more pro-actively as leverage to circumvent such problems. We provide this caveat up front, and also find ourselves repeating it in defence of project implementers, but nevertheless believe our best contribution lies in providing a critical inspection of NRMP. B. HISTORICAL BACKGROUND OF CAMPFIRE 1. Phase I: Initiating CAMPFIRE For the purpose of analysis, CAMPFIRE breaks conveniently into three phases. Having successfully experimented with devolved proprietorship on private land starting in the 1960s, DNPWLM officials attempted to transpose these principles to communal lands from the mid-1970s. Technical wildlife officers in DNPWLM, who saw in the power of localised economic benefit a means of both conserving wildlife and promoting sound land use, initiated the programme, starting in the early 1970s to push through parliament an acceptance that wildlife revenues should be returned to producer communities. In 1989, Appropriate Authority was first given to Districts, and DNPWLM co-opted three NGOs to support their quest. Between 1989 and 1995, the program experienced rapid conceptual growth to a level that is sophisticated even by today's standards. It was coordinated informally, and the balance between financial support and wildlife revenues was well in favour of the latter (there was almost no aid money at this time). Wildlife revenues generally derive from safari hunting, but also from tourism operations. CAMPFIRE was always seen as a long-term programmatic approach that used the fiscal devolution of wildlife revenues to revolutionise rural resource governance. However, from the start the program was viewed with some suspicion, and the greatest threats to it were political. Recognising that political survival depended on 'indigenising' the programme, and that it needed leadership and legitimate representation, the founding group nurtured the CAMPFIRE Association (CA), a body representing RDCs to which Appropriate Authority over wildlife has been given. From CAMPFIRE Association's inaugural workshop at the Park Lane Hotel in 1990, the need to have direct representation by "producer communities3" was recognised. The other key implementing persons, who worked closely with a small technical group at the programme's centre, were the Executive Officers of what were then District Councils and their crucial role and commitment is seldom recognised in the literature. Managerially, a key event was a workshop at Hunyani Hills in which the six partners agreed to a common vision and clear division of responsibility. This diverse group of Government and non-governmental partner organizations were key to the early success of CAMPFIRE. Called the CAMPFIRE Collaborative Group (CCG), this was initially chaired and led by the Department of National Parks and Wildlife Management (DNPWLM). The CCG comprised: • Department of National Parks and Wildlife Management (DNPWLM) in the Ministry of Environment and Tourism, • Ministry of Local Government, Rural and Urban Development (MLGRUD); • The Zimbabwe Trust (ZIMTRUST); • The University of Zimbabwe's Centre for Applied Social Sciences (CASS); • The World Wide Fund for Nature (WWF), and • The CAMPFIRE Association. 3 "Producer communities" is a term coined by the CAMPFIRE programme to refer to those communities in which animals are utilised or tourism occurs. Conceptually, the intention is to intemalise the costs and benefits of wildlife management. NRMPFINALREPORT 29JUL Y2003 19 In 1989 and 1990, "Appropriate Authority" was granted to the twelve 'wildlife' districts, giving legal control of wildlife to District Councils but under the administrative condition that control was further devolved. Thus we have a situation with legislated devolution to RDCs, supported by administrative devolution to producer communities. In fulfilment of their legislated mandate to conserve wildlife throughout the country, DNPWLM replaced the nationalised control of wildlife with conditionalities that wildlife revenues should benefit producer communities directly. The initial conditions attached to appropriate authority status were laid out in a letter by the then Deputy Director of DNPWLM, Mr. George Pangeti. These conditions stated that: • At least 50% of revenues should be devolved to producer communities, • No more than 35% should be used for management, • No more than 15% should be extracted for council overhead. These condition were later (1990) elaborated in the "CAMPFIRE Guidelines", a four page document that set out the "CAMPFIRE Principles" (see Annex P). In these guidelines, DNPWLM set the target of ensuring that fully 80% of wildlife revenues should accrue directly to producer communities. In 1993, the Director of DNPWLM wrote to all RDCs with appropriate authority stipulating this condition and a four-year time frame to achieve it (see Annex P). DNPWLM recognised the power of legislated devolution, but also that this must be extended to producer communities legally. The use of administrative devolution reflected the absence of administrative institutions with legal personality below the RDC, and was always seen as only a stepping-stone to fully legislated devolution. However, DNPWLM's attempt to translate its recognition of the critical importance of devolution to sub-district level into legal institutions, was thwarted. Nevertheless, by persuading key district officials of the importance of linking the benefits from wildlife as closely as possible to the producer communities, and by working closely and mutually with these officials to monitor financial devolution, tremendous progress was made. By 1993, 73% of wildlife revenues were reaching the producer communities, with 66% of these revenues under direct community control (Child, et ai, 1997). The relationship between fiscal devolution and the improving performance of community institutions is now well recognised (e.g., see Child 1993, 1996; ZimTrust 2001,2003) Contrary to the common belief that the program was aimed at wildlife conservation, its protagonists viewed it as primarily about institutional economics and governance. The initial objective was to internalise costs and benefits at the local level, making for more rational and sustainable decision-making. It soon became apparent that grass-roots economic empowerment also had far-ranging political and governance implications. Thus wildlife was perceived not so much in its own right, but as the economic fuel to build a bottom-up system. Having been nationalised for decades, there were few local vested interests in the newly devolved wildlife resource. By devolving economic power to communities, DNPWLM hoped to use this 'new' resource to initiate and experiment with market-like systems for allocating resources to best use. The economic arguments about property rights, exclusion, price and resource allocation figure prominently in the original CAMPFIRE documents (Martin, 1984, 1986). While initially based on this economic and land use rationale, the realisation of the political implication of devolved rural governance was not far behind. Local "proprietorship" became a central tenet of CAMPFIRE, in its three facets: the right to manage; the right to benefit from; and the right to sell. Murphree (1994) articulated these as early as 1990. These principles are encapsulated in three of CAMPFIRE's most important innovations: • Community quota-setting, which implies the right to manage4, • Cash revenue distributionS, which implies both the right to benefit and the right to choose how to use these benefits, and • Community tendering and sale of hunting and lodge concessions6 , which provides the right of disposal or sale. 4 These field-developed tools are now captured neatly in WWF-SARPO manuals. S See Child and Peterson (1991) for a detailed anecdotal description of the first community revenue distribution in Beitbridge. Guruve paid out 2$200 to people in Masoka earlier that this, but this decision was made centrally. See also Child, et aI, ( 1997) NRMPFINALREPORT 20 These tools provided the vehicles for instrumentalising these principles and thereby revolutionised the programme. Cash distribution, in particular, provided a huge impulse, and effectively converted wildlife from a public to a private asset. Improved marketing resulted in a four-fold increase In hunting revenues in the early 1990s. An important aspect of this phase of the programme at this time (1989-1995) was the low level of top-down funding . Some technical support was provided by CASS, ZimTrust and WWF in the form of a few workshops and the availability of technical specialists to communities. From 1993 DNPWLM received minor but critical support from NORAD in the form of a vehicle. Nyaminyami, as the first CAMPFIRE district to be funded, fallured to recover from the centralised organisation and imposed overheads that resulted from this funding. This became an important lesson for the programme, especially as several non-funded districts performed better (e.g. Chipinge, Beitbridge). Moreover, at the time CAMPFIRE was designed, donor-aid was still a rarity in Zimbabwe, and DNPWLM never really considered this possibility. This is why the determination to create wealth by producing wildlife was so strong at all levels of the programme. The key support was the work and effort of a small number of dedicated professionals, the clarity and efficacy of the CAMPFIRE Principles and their devolutionary spirit. Indeed, donor money was seen as largely harmful. In 1989, the same year as the ivory ban and when elephant conservation and trade was receiving heavy publicity, USAID decided to support this innovative, home-grown CBNRM concept, with a pilot initiative in four districts in Matabeleland. USAID provided some USD4.18 million to four districts in Matabeleland, out of a total of USD7.6 million, the remainder of which was used specifically for wildlife management by DNPWLM. As with Nyaminyami, unpublished DNPWLM reports (Martin, pers. comms.) indicated a concern that funding by itself gave districts no discern able performance advantage, and while funding was important for some support functions, there was a lot to learn about how to effectively inject finances into communities without causing as much harm as good. From its inception, CAMPFIRE revenue grew steadily, reaching approximately US2 million by 2000. By the late 1990s, an estimated 90,000 households (630,000 people) were benefiting from CAMPFIRE revenue (Muir-Leresche et ai, 2003). As early as 1991, the collaborating partners recognised the legitimacy and political importance of genuine constituency representation and supported the development of the CAMPFIRE Association (CA), passing to it the role of 'lead agency' in recognition of this legitimacy. While cognisant of its capacity constraints, a founding principle of CAMPFIRE was that "authority is a pre-requisite for responsible management and should not be held out as a reward for it" (SASUSG, 2002). The legal absence of SUb-district bodies corporate resulted again in a compromise, with CAMPFIRE Association legally representing Rural District Councils (RDCs) rather than being directly accountable to producer communities. CA's mission was to promote and facilitate CAMPFIRE by engaging RDCs, their constituent communities and other relevant public and private agencies to implement and support activities conducive to the development of CAMPFIRE (USAID, 2003). . 2. Phase II: Funding, Programme Diversification, CAMPFIRE Association and RDCs 1996 was an important change pOint in the programme. DNPWLM rapidly lost capacity, including several of the early champions of the programme. The CAMPFIRE Association exerted itself as the lead agency, and began to marginalize some of its NGO partners. With considerable suspicion of the part of government as to the motivation of NGOs, and indeed the CAMPFIRE programme as a whole (which was perceived as protected white commercial interests), the expansion of the programme to new areas using USAID financing was important for creating a wider political constituency and ensuring the political sustainability of CAMPFIRE. Thus, 1996 marks the start of significant national funding from USAID, and a desire to include more districts and diversify the programmes economic base from a heavy dependence on wildlife (Le. mainly hunting, but also high-end tourism), Winning political support by "indigenising" the programme had significant costs in terms of the loss of skills that had originally developed CAMPFIRE, and reduced the experience and b See Child and Bond (1994) for an early description of innovations in community-private negotiations; see Child (1995) for an assessment of the impact of marketing innovations on wildlife revenues; and see the WWF manuals for a "how to do it" description. NRMPFINALREPORT 2<)JUL Y2003 21 flexibility to create an innovative funding mechanism that was commercial-orientated and re-enforced devolution. Consequently, the CDF was much more similar to the typical top-down donor fund than it might have been. On the positive side financial support to dedicated CAMPFIRE staff at district level improved natural resource management and facilitated diversification. Unfortunately, it also re-enforced centralisation and a donor-dependency mentality. Whereas the designers of CAMPFIRE wanted to foster a spirit of self-help and production, the temptation for CA and the RDCs was to maintain an easy supply of free money, which shifted incentives towards accessing donor funding. The wish list of 150 project submitted by RDCs in the first year of NRMP II is an indicator of this mentality. Learning how to use grant money effectively was an important part of NRMPII, and is discussed in detail in the text. So are some of the problems caused by this money such as an increased rivalries and jealousies. With a large amount of money to be allocated centrally, leadership was transferred from the more intellectual CCG to the more project- and administratively-oriented Project Execution Committee (PET). In designing the PET, great care was taken to balance the competing forces of economic and technical effectiveness, and political largesse. PET provided for political representation by giving CAMPFIRE Association more power and resources, but ensured that CAMPFIRE's civic and technical proponents retained significant influence. However, CA undermined the role of PET, which became non-functional and an impediment to CDF project implementation and was consequently replaced by the PMT. The CA used their strengthened leadership role to dictate the composition of the PMT, which lost civic and technical capacity and further empowered CA, RDCs and government agencies. In the face of PET non-performance and CA's more dominant position in the program, this revised structure for the project was reluctantly accepted by USAID. After the transition to the PMT, the project continued to directly fund most (except for ART and ZT) CCG members into 2000, and intermittently after that (WWF especially). In an effort to keep the CCG members engaged in the project/programme, USAID made it clear to all CCG members that "the door was always open" for additional CCG funding proposals (even up to 2003). This brought some new members on board (e.g., SAFIRE, FC, MLGPWNH). However, with few exceptions (WWF), most CCG members lost interest in CAMPFIRE when the core funding ceased, a response that was exacerbated (caused?) by CA's negative position on this grouping. In addition, USAID always offered resources to promote/facilitate continued CCG operations/collaboration. However, the CA never pursued this option (as programme lead agency, USAID states that it was their responsibility to initiate this, not USAID's, Loken comments), in a deliberate effort to further marginalize this group and seize tighter control of the programme (and project). This created a leadership crisis in the CAMPFIRE movement. CA was empire-building without the capacity or vision to take on all the roles it envisaged for itself. The CCG, which had created the programme, was sidelined and in any case lost capacity. USAID, which controlled the money, was placed in the invidious position that its decisions were highly influential and was de facto in an uncomfortable leadership position. The net result was that the leadership, maturity and innovativeness of the CAMPFIRE programme was seriously weakened. Given what has transpired in Zimbabwe over the past several years, this transition may have been inevitable, and may have happened anyway with or without the project though perhaps just not as quickly. It was USAID money that allowed CA to become so dominant (a negative) but this money also spread the programme and maintained the CAMPFIRE movement in a time of serious difficulties (very positive). Conceptually, both the PMT and the use of the IC (as opposed to developing long term programmatic capacity) were not the right way to go, but operationally in the light of political events they proved to be invaluable. The project needed a good diligent management entity willing to deal with all of the boring details. The PET/CCG was not doing this, did not want to do this, and (it could be argued) really shouldn't be doing this but attending more important programmatic matters. And, despite all of its problems (structure, potential for conflict of interest, political (equity) considerations, poor consultant pool, poor leadership, etc.), overall, the PMT did an adequate job from the project's perspective, especially for the CDF, which is what the PMT was always most interested in. (acknowledging a lot of extra help from the IC/DA). To purchase political support and broaden its base, there was a strong temptation for CAMPFIRE Association to spread CCDF money and projects equally to as many districts as possible to increase its NRMPFINALREPORT 29JUL.Y2003 22 power base, even where this did not make financial or conservation sense. Despite sacrificing technical efficiency, there is little doubt that the strategy taken by CAMPFIRE Association (led, dynamically, by the late T.N, Maveneke) was crucial for CAMPFIRE's political survival, both nationally and internationally7. However, it also began to divert the NRMP from the principles of CAMPFIRE, with increasing weight being given to more conventional donor 'products' and systems such as centrally funded projects, and rather less on self-production. A noted in the lessons, a better approach, at least retrospectively, would have been to structure grants in the manner of what are now called "social investment funds", where grants are linked to matching contributions and where effective implementation and performance graduates a community to the next level of support. The dangers of this top down funding, and the likelihood of it centralising the programme, were recognised at the time of NRMP design (c.f. Murphree interview). Seeing that the core strength of CAMPFIRE was revenue distribution and control at producer community level, a proposal was made by DNPWLM to gear both the institutional capacity-building and NR investment grants directly to reward and reinforce fiscal devolution. For instance, if $100,000 was devolved to a producer community, this would be matched by $100,000 for investment, $50,000 to purchase appropriate skills through a voucher system (thus also demand-driving service provision) and a similar amount to the respective councils to give them a vested interest in devolution. The donor showed no serious interest in this proposal. Contributing factors may have been the absence of administrative precedent, and the fact that the grant was bi-Iateral and therefore inextricably linked with emerging GOZ reluctance to really support devolution. Whether the reason was USIAD inflexibility, or a judgement on government's reluctance to be innovative, the result is that an important opportunity for experimentation and innovation was lost. There are some suggestions that such an approach might not, in any case have worked, given that the liberalising conditions in which CAMPFIRE was borne, and the strong governmental support provided through DNPWLM, were reversing. At least one service provider was also adamantly opposed to the concept, which would demand-drive rather than supply-drive the programme thereby increasing consumer choice and the pressure for support agencies to respond and perform. 3. Phase III: Winding down of External Support 1998 was a further change point in the programme. Considerable pressure was placed on USAID by anti￾hunting lobbies in the USA, with the result that the programme could no longer explicitly focus on hunting. This supported an internal demand for diversification and a spreading of the programme to non-wildlife districts, which the CDF supported. Diversification was tackled largely through micro-projects including the construction of eco-tourism projects and the sale of veldt products. Despite some efforts by USAID and the IC to encourage this, very little money was spend on market analysis and research, product development and marketing systems. As noted above, after the Mid Term Review PET was replaced by the Project Management Team as the primary mechanism for administering project grants. At the same time, the over-sight of the CCG collapsed, as skills were lost. This changed the nature of the support to CAMPFIRE. The PMT, which was technical, took over the lead of CAMPFIRE by default even though it was not designed for this role. Much of the earlier capacity for long-term vision and technical analYSis was lost. Short-term project financing and constituency-building considerations became important. One result was that investments were based less on technical considerations such as economic returns and more on political considerations such as the spreading of CDF finances to broaden the programmes support base. While, generally, the principle of producer community survived, this was not always the case. In several districts, benefits were spread more widely and lost impact (as the case of Hwange, cited below, demonstrates), while the retention of income by RDCs increased. The balance also shifted significantly from civic to governmental control in the form of Ministries, RDCs and USAID. AI the start of NRMP II, significant funding was provided to CAMPFIRE Units through institutional capacity￾building grants. CAMPFIRE units, which also go by other names, usually comprise an executive officer and assistant (and sometimes resource monitors), are attached to councils and are responsible for monitoring and supporting the CAMPFIRE programme and natural resource management activities in general. They 7 A major influence at this time was international pressure against the consumptive use of wildlife. With some two￾thirds of its income being generated by Zimbabwe's rapidly expanding elephant herds (WWF, 2003), the move to place elephants on CITES appendix I, and therefore to close commercial trade, was a major threat both to the programme and to the incentivisation Qf elephant conservation. NRMPFINAI..REPORT 29JULY2003 23 were provided with a standard set of equipment including a vehicle and computers. Remembering what had happened with Nyaminyami, the evaluation team was concerned that the phase out of grant funding would leave a large central overhead to producer communities, thereby under-mining the principles of CAMPFIRE. In the event, support of these Units has been successful as they continue to provide good support to communities. Moreover, it was not the cost of these units, but increased revenue extraction by RDCs, that was the primary culprit in communities' reduced share of wildlife revenues. At least half of the CDF projects have the potential to be viable, at least in ordinary economic circumstances. Looking to the future, with the imminent phasing out of significant grant funding from USAID, an important question is whether CAMPFIRE Association managed these funds wisely to create economic sustainability both for itself and its constituency. Another important question is whether the diversification policy has worked, although measuring this will be difficult because of the time lags associated with diversification, and the effects of recent macro-economic implosion on both the country's national economy and its tourism sector. We also ask how well CAMPFIRE has stood up to the tests of time, including the difficult economic environment. C. MACRO·ECONOMIC CONDITIONS TO CAMPFIRE One of the driving forces behind the CAMPFIRE programme was the realisation that rapid population growth was over-running wilderness areas in the Sebungwe and Mid-Zambezi Valleys, and that the option to use wildlife was being lost despite wildlife proving to have a comparative economic advantage in semi￾arid rangelands. Zimbabwe's overall population was growing at well over 3% per annum and sufficient formal employment was not being created (Figure 1), forcing people to subsist off the land. CAMPFIRE had to contend with massive population and economic pressures. FIGURE 1: ZIMBABWE POPULATION AND EMPLOYMENT Source: WWF Database 1------ Zimbabwe Populations and Employment I 16.000.000 14.000.000 ... ----- ---.-..... - .... 12.000.000 1:::: . _ _, ._ •• _.-• '. --I •• 6.000.000 _ . __ ::9 __ e_ ..e ____ ... _ _ ___________ 1-Employed l 4.000.000 ... --... --.. --.. . ---------_._-_ .. _-_._---_ .. _-------- .. - . -• .---.--.-.--.--..---.-.. - . -.""FFIf Zimbabwe's general macro-economic circumstances are summarised by GDP data (Figure 2). The liberalising economy in which CAMPFIRE was born, was associated with a steady increase in GDP. However, a return to central planning and control as well as political uncertainty caused a rapid decline in economic performance after 1996, the time when NRMP II was initiated (signed 1994, initiated 1995). This affected all key macro-economic indicators (see Table 1, Figure 2). In the period 1989-1991, the exchange rate was over-valued by about 50%. It was allowed to achieve equilibrium by September 1991, and then floated freely with several crashes (e.g. November 1997; August 1998). The currency was then effectively recontrolled, remaining at ZD38:USD1 in 1999/2000 and ZD55:1 in 2000/2001. This created an informal in foreign currency,. The ZD is estimated to have been over-valued by 10% in 1999,20% in 2000, In 2001, reaching 600-1,000% by 2002 (Muir-Leresche, et ai, 2003). The increasing gap between the offiCial and parallel exchange rate (Table 1) is indicative of the increasing distortion and difficulties of in the Zimbabwean economy. It is of particular relevance to CAMPFIRE because trophy hunting IS SOld. In USD, districts and other official bodies are required to exchange their earnings at the official rates, and thiS represents a huge loss of purchasing power. NRMPFINALREPORT 29JULY2003. 24 FIGURE 2: ZIMBABWE GROSS DOMESTIC PRODUCT (USD BILLION) Source: WWF Database Zimbabwe Gross Domestic Product (US$ Billion) uO.oo .--.-.--.--------- - -- - - --- -------------, 58.00 I---._-.----- ---------.-.--------!i;iil}-- r;rn-------------j 56.00 $4.00 $2.00 1990 1991 1992 1993 1994 1995 TABLE 1: ZIMBABWE: MACRO·EcONOMIC INDICATORS Source: WWF Database; DA for parallel exchange rates 1996 1997 1998 1999 2000 Zimbabwe: Macro Economic Indicators 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 lGop (US$ Billion) $7.90 $6.73 $6.56 $6.84 $7.08 $8.42 $8.20 $5.57 $4.39 GOP Real Growth Rate 2.2% 0.0% -6.8% 4.2% 5.3% -3.2% 8.1% 3.7% 1.6% 1.2% GOP per capital $783 $646 $609 $613 $614 $707 $667 $439 $336 !Z$/US$ exchange rate $2.47 $3.75 $5.11 $6.53 $8.21 $8.72 $10.07 $12.44 24.374 Parallel Exchange rate $9.7 $11.0 $20.7 $37 $39 $331 Inflation (CPI Index) 23% 42% 28% 22% 23% 22% 19% 32% 59% Prime Lending Rate 17% 26% 29% 31% 31% 30% 32% 41% 61% 2000 2001 $4.61 $3.30 -5.5% -6.5% $342 $238 $44.62 $55.07 $468 $2,000 56% 75% 68% 32% The growth in wildlife and wild land has been driven both by the denationalisation and devolution of these resources to landholders, and by the strong economics of this sector, as illustrated in rapid tourism growth trends (Figure 3). The logic of CAMPFIRE was to capture this value to provide direct incentives to rural people to maintain land under wildlife because this was their best land use option. Very early on, CAMPFIRE adopted the principle that the best land use should prevail, and if this happened to be agriculture rather than wildlife, then agriculture should be promoted. Note that these principles are also encapsulated, and almost replicated word for word, by the Government of Zimbabwe's 2002 policy on wildlife on resettled farms (which is printed as a GOZ document but not yet official adopted). NRMPFINALREPORT 29JULY2003 25 FIGURE 3: ZIMBABWE TOURISM: REGIONAL AND INTERNATIONAL ARRIVALS Number of Animals Hunted in Zimbabwe 12,000 8,000 6,000 4,000 2,000 o 1984 1986 1988 1990 1992 1994 1996 1998 2000 FIGURE 4: THE NUMBER OF ANIMALS HUNTED IN ZIMBABWE, 1984 TO 2000 Number of Animals Hunted in Zimbabwe 16,000------------------------: 6,nnlrlJl---- 4, 2, 1984 1986 1988 1990 1992 1994 1996 1998 2000 CAMPFIRE also intended to build on the success of the wildlife sector on private land. Indeed, at the time of the Mid-Term NRMP II review, Zimbabwe's wildlife populations outside national parks had quadrupled in the past fifteen years as indicated by the value of trophy off-take (Booth, 2002). The growth in the number of animals hunted as trophies reflects a rapid conversion of land to wildlife. On the field trip to Save conservancy, for instance, it was immediately obvious that land that was heavily over-grazed by cattle in the 1980s had recovered, and that wildlife populations were much higher, with sable, wildebeest and other species were common replacing the dominance of a cattle· impala monoculture. This strongly vindicates Zimbabwe's conservation two-pronged conservation policy of: • Devolving use-rights to landholders. • Maximising the revenue potential of wildlife. NRMPFINALREPORT 29JULY2003 26 CAMPFIRE was the response to this policy success, replicating its intention in Communal Areas. Economic research and the conversion of land from livestock to wildlife demonstrated that, where the economy was not distorted, wildlife was a highly competitive form of land use in semi-arid regions (Natural Regions IV and VB). The main concern was the heavy long-term subsidization of the agricultural sector and the implicit taxation of wildlife. Bond (2002) found the subsidization of livelihoods, and not only agriculture, to be the over-powering factor driving the rapid migration of people into unsuitable agro-environments in the 1990s. Bond concluded that people were not flooding into new areas to farm, but to gain access to the many free services provided by government and an open access natural resource property regime - schooling, clinics, fertilizer and agricultural inputs, food, and timber and grass for building. Note that where there may have been some control of access to resources by local people who had developed institutional systems over many years, new settlers often over-whelmed the original inhabitants and seldom respected their resource management systems. In-migration was characterised by an extractive, frontier economic mentality, and did a great deal of harm to the environment (Derman, 1990). D. LEGAL AND POLICY FRAMEWORK CAMPFIRE rests on a number of policy and legal instruments that need to be read together to gain a full understanding of the allocation of rights and responsibilities for natural resource management. The CAMPFIRE approach is based on the sustainable use of natural resources. This principle is articulated in the 1989 Policy for Wildlife, which consolidated into official policy the thinking that had developed in the DNPWLM concerning wildlife use, proprietorship and incentives for sustainable management. Apart from commitments to protected areas and the preservation of endangered species, the policy also contained the following statement, which underlines the economic foundation of the DNPWLM's approach to conservation: "Outside the Parks and Wild Life Estate, government views wildlife as a resource capable of complementing domestic livestock and will favour neither one above the other in the development of the country. It will rather aI/ow economic processes to determine the outcome of competition." (Government of Zimbabwe, 1989: 7) CAMPFIRE is also based on the principle that landholders who live close to the resource and suffer the costs of living with wildlife should be those who benefit from sustainable management. Thus the 1975 Parks and Wildlife Act gives freehold farmers "Appropriate Authority" (AA) or custodianship over wildlife, fish and plants". The Act effectively makes freehold farms and ranches into proprietory units for wildlife management. Farmers are able to make nearly all of the significant management decisions over the use of wildlife. These include the right to hunt, to allow someone else to hunt, the right to buy and sell game, and to carry out trophy hunting. The Act allows for the exemption of specially protected species and for Government to impose restriction orders in cases of flagrant abuse. As described by Child G. (1995), the high standard of conservation on private land was based on the principles that peer-based control was cheaper and far more effective than the heavy hand of government. From as early as the Natural Resources Act of 1941, landholder communities (legislated as "Intensive Conservation Area Committees") had the legal recourse to control any unsustainable land use practices within their area (e.g. over-grazing, timber cutting, soil erosion) and were backed up by experienced government technicians. The Parks and Wild Life Act took advantage of this elegant and effective locally based institutional mechanism, rather than government policing, as the primary method of control, with remarkable results, as Child, G (1995, p87-88) describes in far greater detail. The reason we mention this here is that, should appropriate authority be devolved to producer community level, this should be complemented by a similar peer-based controlling mechanism. The devolution of rights over wildlife to freehold farmers led to the development of a multi-million US dollar wildlife industry on freehold farmland and improved conservation of wildlife. Following this success, the DNPWLM aimed to establish the same incentive-based approach to conservation on communal land. In 1982 an amendment to the Parks and Wild Life Act gave the Minister power to designate Rural District 8 Vincent and Thomas (1960) in their classical agro-ecological survey of Zimbabwe, define five natural regions, of which natural regions 4 and 5 are unsuited to rainfed agriculture and are suited only to extensive or very extensive rangeland production systems. NRMPFINALREPORT 29JULY2003 27 Councils as Appropriate Authorities over wildlife on communal land. If an RDC is designated as an AA it gains the same rights over wildlife as the owners of freehold farms. Martin (2003) suggests the Zimbabwean legislation is the only example in southern Africa of the full devolution of authority over wildlife to entities outside the state. However, devolution of authority over wildlife on communal land stops at the level of the RDCs, which are government bodies. The original intention of policy planners in DNPWLM had been to extend rights over wildlife producer communities i.e. those on whose land the wildlife was found and those who suffered the costs of living with wildlife. The early CAMPFIRE documents (Martin, 1984, 1986) mentioned the concept of the "village company". The planners had to compromise in this objective due to political constraints and were unable to legislate for devolution of authority over wildlife to levels below the district. Policy guidelines were issued stating that RDCs were expected to distribute a percentage of income derived from wildlife use to producer communities and to allow these communities to be responsible for a number of wildlife management activities (Annex P). Because of the existing administrative system of local government, producer communities had to be represented by Ward Development Committees (WADCOs) and Village Development Committees (VIDCOs), which had no legal personality but were advisory bodies to Councils. This is the root cause of the problem that this evaluation detected in the structure of the CAMPFIRE Association. A number of other laws give the RDCs considerable control over natural resources. The Natural Rural Resources Act provides for Councils to establish conservation committees and these committees are able to make bylaws on the use of natural resources, including wildlife. The Communal Land Act vests ownership of communal land in the State and the administration of communal land in the hands of the RDCs. The Rural District Council Act gives the councils power to take measures to conserve natural resources. permit grazing and cultivation, develop land use plans and make bylaws for the protection of natural resources. The councils may issue permits for catching fish, hunting, cutting firewood, cutting grass and collecting honey. The Forest Act allocates large areas of former communal land to the Forestry Commission, which leases timber. hunting and photographic tourism concessions. The Commission has adopted the policy of encouraging RDCs to share 15% of their income from timber royalties with local communities. The Communal Land Forest Produce Act restricts the use of forest products to "own use" and excludes use of products from protected forest areas and areas where a license to cut trees has been granted to others. The use of certain trees is restricted. According to Chitsike (2000:11) under the Act, "without a permit or license, virtually any use of woodland is illegal". The Environmental Management Act aims to establish a general legal foundation for all environmental laws based on sustainable development and addresses inconsistencies, overlaps and duplication In environmental and natural resource legislation. The Act contains limited references to devolution and decentralisation and does not provide for empowering sub-district levels. The Traditional Leaders Act provides for Ward and Village Assemblies that would "consider and resolve" all issues relating to land, water and other natural resources. This statement is somewhat ambiguous with regard to actual decision-making powers of the Assemblies. Further, the Act does not provide land rights to the Assemblies and it does not give them any legal status beyond being sub-committees of council. In contrast to other natural resource legislation, the Water Act makes provision for the establishment of Catchment Councils, which would be responsible for water management in the river systems for which they are established. The Act provides for Catchment Councils to be a body corporate with a legal personality. The Catchment Councils are able to delegate functions to Sub-Catchment Councils that are able to levy rates and fees and open an account in which the income can be held. In summary, with the exception of the Water Act, natural resource legislation in Zimbabwe concentrates considerable power in the hands of Rural District Councils. Policy guidelines in the wildlife sector attempt to promote the "administrative devolution" of some functions and decision-making to sub-district levels. However, RDCs cannot legally be forced to apply the policy guidelines although DNPWLM has stated in the past that it would withdraw Appropriate Authority status if councils did not comply. The policy and legislative framework within which CAMPFIRE operates creates numerous local level institutions that operate in parallel, have overlapping functions and compete with each other for power and access to financial resources. NRMPFINALREPOR r 29)1 JLY2003 28 E. USAID SUPPORT (NRMP) TO CAMPFIRE 1. NRMP I (1989·1994) USAID has been a partner with the Government of Zimbabwe (GOZ) in supporting CAMPFIRE since 1989 when the Agency proposed a grant as the Zimbabwe country-component of a regional program including Zimbabwe, Zambia and Botswana under the Initiative for Southern Africa (ISA) Natural Resources Management Project (690-0251), commonly referred to as NRMP I (1989-94). Project identification, preparation and planning was undertaken by USAID with substantial assistance from the DNPWLM, CASS and the Zimbabwe Trust (ZimTrust), the latter two agencies having been invited into the process at the request of the Department. NRMP I, with an effective implementation date of 15 September, 1989, was fully financed by USAID through an agreement with the GOZ which provided for sub-grants to ZimTrust and CASSo USAID (Loken, comments) states that this was "designed as a pilot initiative to test the innovative/revolutionary [CAMPFIRE] hypothesis on a limited scale, before committing more substantial USAID resources". There were four target districts, namely Tsholotsho, Hwange, Bulilima Mangwe and Binga in Matabeleland, and wildlife management by DNPWLM was also supported in the Hwange-Matetsi complex. Sub-activities undertaken have been in: • Wildlife management, • Institutional and community development, • Community development involving women, and • Education and training. The total obligated project grant for the Zimbabwe component under NRMP I (690-0251.13) was USD 7,600,000 scheduled for a 5-year duration. The sub-grant obligated by USAID to the ZimTrust for implementation was USD1,541 ,000. The Trust was also responsible for receipt and control of USD1 ,400,000 allocated for district-level infrastructure and capital requirements. Although the first USAID project was limited to the four sub-districts under NRMP I, the CAMPFIRE program was already very active in a dozen rural districts and had begun activities in another dozen. By that time the CAMPFIRE Programme had demonstrated notable accomplishments. Two of the most successful CAMPFIRE projects Mahenye and Beitbridge had received no external donor help. Under NRMP I, anticipated project results of increasing community involvement and rationalizing use of fragile lands had been partially achieved, and the potential for nationwide impact had also been convincingly demonstrated through project activities and efforts focused in the four pilot areas. The mid-term evaluation of the Zimbabwe Natural Resources Management Project I, dated January 1995, generally supported the programme although it cited a number of problems and issues related to the financial and administrative aspects of the Project where there were obviously frustrations between USAID and implementing agencies over the complexities of complying with USAID's systems. They included: • Concerns about unpaid vouchers and the fear that the experience of working with USAID might make some partners "fight shy" of further involvement; • The belief, by some, that USAID rules and regulations are more complicated and burdensome than from other donors; • Allegations that queries by USAID of implementing agency compliance with USAID rules "seriously damaged the morale of those agencies, encumbered them with incremental staff they were ill￾equipped to afford, and delayed the completion of tasks." (p. 35) • USAID complaints that participating agencies were tardy in submission of vouchers or submitted incomplete vouchers causing them to be "bounced back" to the submitting organization. In spite of these administrative conflicts, NRMP I was seen as successful and USAID followed up on the successes of NRMP I by continuing and expanding its participation in CAMPFIRE. The successor program (NRMP II) did not restrict its interest to a limited number of districts but instead embraced CAMPFIRE in its totality, freeing up its contribution to be used wherever the programme was active. Moreover, USAID intended to work with CAMPFIRE to expand beyond the area of wildlife conservation and use, to the conservation and use of range lands, forest resources and eyen to mining. Where the Appropriate Authority had been given to RDCs, the NRMP II funding was to follow; with ICB grants to 23 RDCs. Accordingly, USAID considerably expanded the funds it was to contribute to CAMPFIRE over the next four years. At the same time, USAID NRMPFINAl.REPOR r 29JULY2003 29 changed its operational approach to include an Institutional Contractor (IC), so that funds could move more quickly and be more responsive to the needs of CAMPFIRE and the RDCs. A US Company, Development Associates (DA) was engaged as the institutional contractor. USAID (Loken, comments) notes that the IC/DA was specifically engaged to address the problems with compliance with USAID procedures identified in the four evaluation bullets drawn from the NRMP I evaluation (Pangeti et, al report). Others have suggested that the IC was expensive, did not contribute to the long-term capacity and sustainability of the programme, and may have contributed to the marginalisation of the NGOs. In the second phase of NRMP, which was authorized on September 30, 1994, the envisaged USAID contribution was USD 20.5 million, which included USD 16 million in bilateral funds and USD 4.5 million in regional funds. This brought the total USAID contribution over the life of the project to USD 28,100,000 with a 25% GOZ contribution in kind, estimated at USD 9,400,000. 2. NRMP II Project Design: Goal, Purpose and Outputs In September 1994, NRMP II had two goals: • A multi-country regional goal of regional cooperation in promoting natural resources activities which will contribute to the sustainable development of communities on lands that are marginally sustainable for agriculture; and • A country-level goal to use natural resources management to develop economically sustainable communities on lands marginally suitable for agriculture. Under Phase II, the two country-level purposes were: • To develop community-based programs to increase income while sustaining natural resources; and • To improve local capabilities to protect the resource base. The achievement of the project purpose was to be measured by achieving the following end of project status: • Natural resource management programs underway in 23 districts; • Revenue distributions made to all participating communities; • Stable wildlife populations in targeted areas; • Flexible and analytic responses to changing situations, formulated by CAMPFIRE program implementers. The planned outputs of NRMP II in September 1994 were: 1. Community-level resource management capacity institutionalised; 2. Strengthened CAMPFIRE Association capable of supporting its members in community-based natural resources management programs; 3. Knowledge of NRM increased, especially socio-economic variables affecting sustainability of community programs; 4. Education and outreach expanded; 5. Policy analysis completed; and 6. Networking and communication systems in place. These outputs were to be achieved through four project components: I. Community-based resource management and utilization II. Planning and applied research III. Wildlife and natural resources conservation; and IV. Regional communications and information exchange. NRMPFINALREPORT 29JUL Y2003 30 3. Direct and Indirect Project Beneficiaries As far as beneficiaries are concerned, NRMP II saw congruence with the overall CAMPFIRE programme, which aimed to empower communal farmers to sustainably manage and conserve the wildlife and natural resources through the development of locality institutions. The primary beneficiaries under the CAMPFIRE program have always been households at community (village and ward) level. Under NRMP II, these communities were the intended ultimate beneficiaries. However, the RDCs, the CA and the CAMPFIRE Service Providers became the vehicles to reach the communities. As the latter became direct and immediate beneficiaries under NRMP II, a major share of project resources went towards building capacity and providing facilities in these institutions with the purpose of reaching the ultimate beneficiaries. 4. NRMP II Implementation Strategy prior to the 1998 Mid-Term Evaluation Under NRMP Phase II the implementation approach was to channel the majority of funds through an institutional contractor selected through competitive means, which was to serve as a secretariat for implementing agencies, as a liaison/buffer between project agencies and with USAID, and to provide capacity-building and training. However, USAID was to maintain direct relationships with, and disbursements to, DNPWLM, the Ministry of Local Government Public Works and National Housing (MLGPWNH), Department of Natural Resources and the Forestry Commission. The strategy also called for: • Expansion of services to many more districts, • Financial, technical and training support to the CAMPFIRE Association, to strengthen its role as the lead CAMPFIRE agency, • Introduction of an adaptive management and coordination system to better service the large number of CAMPFIRE stakeholders and implementing agencies, • Establishment of the CAMPFIRE Development Fund (CDF) to fund projects designed to strengthen the capacity of RDCs to service CAMPFIRE and support community initiatives to improve natural resources management, and the • Appointment of an Institutional Contractor (IC) to administer the USAID funds allocated to support CAMPFIRE. Stakeholders organized themselves to implement these initiatives starting in 1995/6. Subsequently, as the Programme evolved, CAMPFIRE responded to other priorities such as the need to diversify the Programme from consumptive to non-consumptive natural resource utilization activities. 5. Revised Implementation Strategy after the 1998 Mid-Term Evaluation In 1998, following the 1997 CITES Conference in Harare and the Project's Mid-Term Evaluation, and in agreement with USAID's (then) new re-engineering precepts, the Project was modified to its current and final form as the Mission's Strategic Objective One (S01) is, "strengthened NRM for the sustainable development of CAMPFIRE areas." Replacement of PET by PMT An important change following the Mid-Term Review was the replacement of the PET by the PMT. Membership of these coordinating mechanisms were as follows (Table 2): TABLE 2: MEMBERSHIP OF PET AND PMT Membership Project Execution Team Project Management Team GOZ DNPWLM, MLGRUD DNPWLM, MLGPWNH CCG/ Civil Society CASS, ZimTrust, WWF, others Donor/ManaQement USAID/DA USAID,DA Secretariat CA CA Beneficiaries RDCs Up until 1994 the CCG functioned as a coordination mechanism. It was an important strength in the program ("coalition of support agencies"). PET was designed to build on the strengths of the CAMPFIRE Collaborative Group (e.g. a powerful dynamic of innovation, peer review, cooperation and constructive criticism), but to formalize this alliance in a balanced relationship between technical. political, civic and NRMPFINALREPORT 29JUL Y2003 31 government forces, to cope with large amounts of project money. This never worked. The Mid-Term Review recommended the replacement of PET by PMT. This reflected the political pressure of the time, including the CA desire to dominate NRMPII finances, GOZ's desire to marginalize NGOs, and the need to push through COF projects. However, it also weakened and unbalanced programme leadership. It appears that the Initial intentions of the PET, and the solutions to its under-performance, were diagnosed as structural, when they were in fact related to personalities. Reasons we have been given are that: • PET was badly led by CA, became bureaucratic and unproductive, and deliberately marginalized NGOs • Partners therefore did not take it seriously (poor leadership led to reluctant support), and was attended (poorly) by junior CCG but seldom the bosses and • Partners were in any case losing technical capacity. In this period there was considerable competition over roles and the limelight, particularly between ZimTrust and the CA, which "mis-understood its role and wanted to be everywhere", leading the CA Director to state: ''To hell with NGOs". The fight over NRMP II resources was often highly antagonistic. Part of the leadership problem relates to the loss of key CCG staff at this time. However, much relates to a power struggle between CA and the NGOs. Thus, CA probably had little intention of building PET into a collaborating team, and did not have the capacity to lead PET effectively, and may have deliberately discouraged participation. Some CCG members openly flouted CA leadership, were perceived as frustrating efforts to facilitate COF projects, and did not always provide technical services as required or demanded, to quote: "they are not in this year's work programme". The pressure to 'indigenise' may have also been used by CA and ROCs to gain dominance over resource allocation. This antagonism was a factor in the shift to using consultants rather then CCG members. However our analysis is that the problem lay not in PET's structure, but in the way it was led and managed (indeed, almost ceased to function), CA's desire to control the programme and its resources, avoid over-sight (e.g., get rid of the NGOs who were constantly pestering them, speaking on their behalf, interfering in their business, etc.) and the acrimony of the period. PMT was an improvement measured in terms of the delivery of COF projects, and was not Intended to replace the CCG. Nevertheless, it de facto lead the programme, but at the time of the evaluation lacked the capacity to do so, with key sustainability and other issues not being addressed. NRMPFINALREPORT 29JULY2003 32 6. Responsiveness of NRMP II to Recommendations of the Mid-Term Evaluation 1.1 Recommendation USAIDIZimbabwe should continue to support the CAMPFIRE movement, with some modification to NRMP /I and with additional funding through FY2002. • Continue direct contract support to the GOZ and IC partner, and continue grant support at a reduced level to the CA. • Maintain initial direct support to NGO Partners (CAMPFIRE Service Providers - CSPs). Subsequently, phase down direct grants to NGO partners, and, instead, encourage them to continue to access support through joint proposals with the CA Secretariat and its membership (Rural Councils) for funding under the Community Oevelopment Fund (COF) or other sources. Evaluation Comments Support was continued until September 2003, with no-cost extensions made possible by savings on expenditure and exchange rate gains. USAID continued its support to DA and CA through 2003, although the CA grant was outside the DA contract for the 2000-2003 period. USAID funded small programs with Forestry Commission (FC), Department of Natural Resources (DNR) and Local Government (MLGPWNH) soon after the Mid-Term Evaluation, and then discontinued them in late 2002 in response to the new United States Government (USG) policy not to support state agencies in Zimbabwe. Grants to CASS, ZimTrust and SAFIRE terminated in late 1999 / early 2000. Some joint (the CSP with CA general endorsement) proposals were developed and funded - WWF for specific M&E functions, aerial surveys and other agreed upon tasks; and SAFIRE for specific community trust development work. Communities requested many of the aforementioned services - for the development of community trusts and in some cases for WWF training. 1.2 Recommendation Modify the CDF to enhance CAMPFIRE sustainability and diversify the sources of revenue from sustainable natural resource use. Facilitate review and decisions regarding current proposals under active consideration; Open up access to COF monies by eliminating the distinction between capacity-building and infrastructure grants; Revise COF proposal criteria to encourage (1) actions at the sub-Rural level and (2) economically viable natural resource enterprises through pre-investment feasibility studies and capital investments in the resource base; Have USAIO and the MLGPWNH begin to involve neutral peer reviewers in decision-making about the allocation of COF monies; Oevelop a decision strategy for extending the COF beyond the current project completion date. Transform the COF into a longer term financing mechanism, such as a sinking fund, that would extend NRMP " funds beyond the potential ending of USAIO support in FY2000 or 2002. This would provide new, as well as ongoing initiatives [within] the necessary time to become technically sound and sustainable. Use the COFlfollow-on trust or endowment mechanism to facilitate and provide an enabling context especially at the vii/age and ward levels, for the development of commercially viable NRM enterprises, identified through the participatory planning process. NRMPFINALREPORT 29JUL Y2003 33 Evaluation Comments According to DA (2003a), the thrust of these recommendations was to expedite CDF implementation, move into more infrastructural projects (as opposed to ROC training/capacity building activities that had been the main focus of earlier support), and broaden the range of recipients and the types of NRM activities being supported. It was also evident that the CDF would terminate in a couple of years, and that steps should quickly be taken to generate additional funding for NRM projects in the future. Most of the recommendations were implemented except two: CDF was not turned into a trust or endowment fund; and neutral peer reviewers were not involved to any significant extent in decision-making about the allocation of CDF monies. Comments on developments following these recommendations are as follows: Shortly after the stakeholders' workshop to discuss the results of the 1998 Mid-Term Evaluation, USAID amended the NRMP to reflect the new management framework agreed upon at the workshop. The framework sought to simplify' the CDF and NRMP decision-making process, involve CA and key government stakeholders more significantly in management decisions pertaining to the NRMP (and, in particular, to the CDF). This new arrangement (details found in the NRMP Grant/Amendments file) did expedite the CDF review process by disbanding the PET, a relatively powerless CCG Committee chaired ineffectively by CA's Deputy Director, and replaced it with a smaller but higher-level group (the PMT) chaired by the CA Board. The PMT soon began to hold regular monthly meetings to approve or reject projects, and this practice continued for most of the period. The CDF proposal criteria for infrastructure projects were revised/finalized in late 1998 and RDCs were trained again in proposal development. The revised criteria gave more weight to sub-district participation in project development and to pre-investment analysis (particularly financial/economic). The CDF did achieve much more diversification in its portfolio, with particular emphasis on eco-tourism and other forms on non-consumptive NRM activities (see Annex O). Response to Mid Term Review In direct response to the specific recommendations of the Mid-Term Review: • Processing of proposals was considerably speeded up. CA played a stronger leadership role in the PMT. A more central role of the IC/USAID in decision making, and of the IC In CDF management, was accepted. The IC recruited technical capacity in tourism and engineering. However, it was still slow, with a lack of responsiveness by RDCs and complex procedures both being blamed. To quote from the field trip report, the large project in Gokwe North took 30 months, the small grant took 15 months, and the Nyanga Kairezi grant took two and a half years. Here DA suggests that ROC capacity was a major constraint, and indeed some RDCs took months to follow up on recommendations, whereas RDCs feel procedures were complex. Whatever the case, it appears that there was not enough communication between the PMT and RDCs to facilitate project implementation. • DA (2003b) confirms that the distinction between capacity-building and infrastructure grants was removed, and agreement was formally reached to concentrate the remaining CDF funds on infrastructure projects (DA, comments). • Latterly, most projects were implemented at the sub-district level (see DA, 2003b), and economic and technical criteria were improved. While there was nothing conceptually wrong with these criteria, requirements were excessive, were driven more by USAID requirements than the requirements of communities and they were not particularly user friendly, with some difficulties experienced in managing the balance between the use of rigorous criteria and implementation flexibility. Table 29 summaries the large CDF grants made, all of which were after 1998, showing that at least half were done at sub-district level. Table 3 summarises the technical viability of these projects. • Neutral peer-review was not incorporated into the decision process, although it was used on specific issues. The process could have benefited from more private sector input/experience, especially for tourism projects, but there was resistance to this from CA. • It would have been better to have more proposals as private sector - community joint ventures with integrated private sector input in design and implementation (but RDCs and CA discouraged this, although field visits suggested it is what communities wanted). • No strategy for extending CDF beyond NRMP II was agreed. • The CDF was not transformed into a longer-term financing mechanism. USAID rejected this recommendation of the Mid Term Review (Loken, comments). Our observation is that this should have been included in the original project design. NRMI'FINA I.REI'OR r 2'>Jl i\.Y2003 34 Comments on the Sustainability of Large CDF Projects Only two eco-tourism projects stand out as "highly probable" in terms of sustainability. The requisite infrastructure for these projects has been completed and the projects are already operational. Both projects are situated in the Eastern Highlands and they attract local tourists as well as the few regional and international tourists who still visit Zimbabwe. Of the five projects that the team considers as being in the "probable" category, two are eco-tourism projects and three involve fisheries. Five ecotourism projects could possibly be sustainable, but these have not yet been completed, and their viability will depend largely on significant improvements in the political and economic environment in the country. The viability of four projects (mopane worm harvesting and processing, wildlife management, ecotourism and electric fencing) is doubtful. The overall summary is as follows: Probability of Success Projects Highly probable = 2 ecotourism 2 ecotourism 3 fisheries Probable = Possible Likely Difficut Doubtful = 5 ecotourism = 2 conservation = 1 rehabilitation = 1 mopane worm harvesting and processing 1 wildlife management (rehabilitate ranch) 1 ecotourism 1 electric fencing The following notes pertain to the CDF (discussed in more detail in section IV.A): • Large CDF projects were selected equitably, with one viable project in each district. • Viability was defined in terms of sustainability, rather than return on investment. • Over time, grants (Table 28) were increasingly developed at local level, and with more involvement of private partners. Performance improved. • Monitoring of financial and economic impact was difficult to build into the process since impact extend well beyond the life of the Project. • Expectations of viability were unrealistic. Diversification through eco-tourism (Le. lower-middle market) was not yet well developed even by the private sector, but were popularly requested by communities (albeit, often with unrealistic expectations) • With a focus on delivering projects at local level, the importance of the central functions of marketing and product development was recognised but not sufficiently pursued, with CA being unresponsive to this potential and important role. Thus, no marketing / booking system was developed in support of site-level ecotourism investments. • Timing was unfortunate, as these developments were completed after tourism had collapsed. In retrospect, it might well have been better to set up a (soft) loan fund, whereby communities and the private sector could submit joint proposals to fund diversification investments (see recommendations). NRt-1Pf'INAI.REI'ORT 35 TABLE 3: ASSESSMENT OF VIABILITY AND LEVEL OF CDF PROJECTS District Type of project Viability IMPLEMENTER (ROC, Trust Private Partner) Mudzi. Nyatana Wilderness Doubtful. 3 Dislrict Trust (7 wards) Rushinga, UMP Chiredzi Veld Fire Management Conservation, likely ROC Chimanimam Chalets, ornithology Highly probable Trust Mazowe Ecotourism camping Marginal Trust Muzarabani UQgrade with chalets Probable ROC; Private partner Chipinge Veld Fire Management Conservation, likely Local, with ROC support Nyanga Fishing chalets Highly probable Trust, with Private sector and ROC support Hurungwe Small scale fencing/revolving fund Doubtful. ROC and Community Binga Electric fences Conservation, likely ROC and Communities Gokwe North Vetd Fire Management Conservation, likely ROC Hwange Fishing camp Probable ROC -- - Beitbridge Fisheries Probable Community groups Bulilima and Amacimbi (mopane worms harvesting and Doubtful 2 ROCs and 2 Trusts Mangwe processingl Umzingwane Mtshabezi VaUey_ Ecotourism Doubtful ROC and Community_Trust Guruve Ecotourism project Probable ROC and 2 Trusts I Gwanda Rehabilitation of Ooddieburn Manyoli Difficult ROC ranches Bindura Ecotourism campsite at Paradise Pools Possible ROC and Trust Matobo Ntunjambili caves ecotourism Possible ROC and Trust Mwenezi Manyuchi fishery Probable ROC and Commun!h'Trust Goromonzi Ngomakurira Ecotourism day centre Possible ROC and Trust Manyame Mayambara lodges and conference centre Possible but difficult ROC, Trust and Private Partner Peer Review As noted by DA (2003a), the PMT only occasionally used neutral, peer reviews to help decisions on CDF funding allocations. More often than not, USAID or DA prompted such peer reviews after expressing strong reservations on the merits of some projects being reviewed by the PMT (e.g. Binga Fencing, Hwedza honey processing proposal, proposed private sector partnership agreements for Chimanimani and Manyame eco￾tourism projects, Hurungwe and Nyaminyami fencing projects). The inclusion of peer review and experience from the private sector might have had a strong positive influence. Sustainability DA (2003a) confirms that neither CA nor other GOZ representatives on the PMT seemed very concerned about the longer-term sustainability of the CDF. While the other CCG members were interested in establishing a National Trust Fund, CA resisted this (section VII.A). USAID decided against this in the NRMP II design, and also when it was raised by the Mid term Review when the time horizon was too short (Loken comments» 1 .3. Recommendation Simplify NRMP /I administration by phaSing out PET and implementing a series of key project management changes. In the short-term, eliminate PET and incorporate its subcommittees into the CCG. Refocus COF financial management and other NRMP 1/ management responsibilities more tightly within the IC and USAIO S01 team. Reduce the number and frequency of project planning and implementation meetings. For broader project issues, including the impending NRMP 1/ and Mission graduation, the core management team should work closely with the NRMP /I Bilateral Review Committee (BRC). Include new partners and develop a revised Plan of Operations for the second half of the project. NRMPFINALREPORT 29JUL Y2003 36 Evaluation Comments Eliminate PET The PMT replaced the PET. The PET sub-committees were generally not effective, except the M&E Committee chaired by WWF at CA's request. This committee developed an Monitoring & Evaluation Plan and Implemented most aspects of it. This exception was due in part to WWF's and USAID/DA's interest in, and need for, information on the impact of CAMPFIRE/NRMP for reporting purposes. Refocus COF witl7in IC and USAIO S01 The recommendation to refocus CDF financial management more tightly within the IC and USAID S01 team was followed. The PMT held regular meetings and approved more projects in the 1999 / 2000 period under the revised CDF criteria (e.g. more user friendly small projects window, same criteria but reduced requirement for documentation). Note that NRMP stopped reviewing proposals for large projects in 2001, but continued reviewing and approving small projects right up to the end of 2002. Reduce Meeting Frequencies The number of CSP joint planning meetings was reduced from four to three meetings per year starting in late 1998. By the year 2001 , CA discontinued all annual and tri-annual planning/progress review meetings of the CSPs. The initial rationale for having fewer meetings was sound (less paperwork / more action), but the cessation of virtually all meetings seemed extreme. While some CSPs may have lost interest as their DA grants expired, CA made little effort to incorporate them in CAMPFIRE plans and seemed to feel that there was little need for their services. Interviews with several CSPs suggest that they were deliberately marginalized by the CA at this time. This reflects badly on the capacity of CA to source and maintain productive partnerships, and is one consequence of excessive dependence on a single large donor, coupled with competition for a large amount of financial resources in the form of free money (Le. grants). In this manner, unfortunately, the NRMP II Project inadvertently weakened the programmatic focus of CAMPFIRE. The following negative consequences can be attributed to this shift: (a) The creative tension between NGOs, the CA, and government was lost, with the balance shifting too far towards administrative and government agencies. (b) The diverse strengths of the CCG, an important contributing factor in the conceptual power of CAMPFIRE, was lost. (c) The opportunity for productive critiCism and peer review was reduced (noting that CCG members also lost capacity). (d) The more mechanical process of project management replaced the ethos and mechanism of adaptive management (e) The innovative capacity of the program was lost, with few major innovations during NRMP II, which had a far more mechanical focus on roll-out and grant management. (f) The inte"ectual and financial portfolio of the CAMPFIRE program was narrowed. As noted, the negative points d-f were not a direct fault of NRMP, although the financial empowerment of CA may have contributed to the deliberate/continual CA obstruction of the CCG. Regarding point f, there was some broadening of the programme (e.g., inclusion of Forestry Commission, SAFIRE). However, the overall impact of the marginalisation of the CCG was (1) a loss of intellectual capacity and (2) a reduction in their desire/capacity to source additional donor funds to support CAMPFIRE, thus broadening the funding portfolio. Bilateral Review Committee (BRC) The BRC never met, as the PMT did not feel that there was ever a need for their intervention in the NRMP. The CA, however, has not yet dealt in any significant way with the broader strategic issues facing CAMPFIRE. So this function, envisioned for the BRC, was not undertaken. DA (2003) suggests that CA's development of its SGI (developed with NRMP funded external technical assistance) may be a step in the right direction, but this is not the opinion of the evaluation (see section VII B). New Partners The 1998 NRMP Grant Amendment opened the door for funding new NGO and GOZ partners over the remainder of the Project, and new partners such as SAFIRE, FC, DNR, and local government took up some NRMI'FINAI.REPORT 29JUI.Y2003 37 NRMP implementation functions. However, the extent of uptake was limited as the government agencies generally lacked the capacity to acquire USAID funds and effectively phased themselves out in 2000 (although funding was available up to May 2D02), and SAFIRE was only directly funded for one year (for reasons of underperformance). NGOs (except WWF) did not make full use of the funding for reasons discussed above. Supporl Mechanisms The CAMPFIRE program appeared to lose its strategic and programmatic focus at the time of the Mid-Term Evaluation: • The simultaneous concentration of responsibility and finances within the CA nexus, while tightening DAIUSAID S01 control, is internally inconsistent. The simultaneous empowerment and then hamstringing of the CA tends to neutralise the intention of this objective? • The project de-emphasised long-term institutional support, and emphasised the use of contracted service providers . The role of several CCG members changed from long-term partners into short-term support provision. This also created a conflict of interest for CCG members: they were competing for funding, and also participating in the adjudication panel. For reasons of conflict of interest, WWF for instance, was happy to withdraw from PET/PMT. Diversity and technical capacity of the CCG/PET was reduced, with the PMT being dominated by government agencies (DNPWLM, MLG, USAID) and by the CA membership. At the very time that greater entrepreneurial and commercial capacity and judgment was required, it was lost from the decision-making process. While DA was able to provide some commercial advice, this placed DA in the difficult position of having to argue against non-viable projects, that the rest of the members were happy to approve. • According to DA (comments) a conscious decision was not made in 1998 to convert CCG to Service Providers. Although the formal name change did occur at the time, it was intended that the CCG continue and concentrate on broader Programme issues. However, the notion that CCG agency were also to be Service Providers was built into the Project Design for NRMP II and was articulated by project implementers from the beginning. CA was to be the lead organization helping the RDCs to obtain better and more relevant CCG services. CCG inputs were to serve these interests and the CCG was not to use the NRMP to implement their own institutional agenda (a common perception of CAlRDCs and some key GOZ Ministries). Ideally, NRMP funds were to be apportioned to only those CSPs that responded to on-the-ground needs identified by RDCs/CA. We agree with the IC that this was good in theory, at least concerning the delivery of specific services, especially as the performance of several of the NGOs was declining at the time. However, it was difficult to put into practice, and could have been more responsive to some worthwhile CCG initiatives (DA comments). In terms of long-term support, we concur with DA that a partnership approach was, in retrospect, more sensible. Indeed, one of the lessons and challenges presented by NRMP II is how to retain long-term programmatic support without losing quality. We see no option but to recruit and rely on people who are profeSSionals in their field. In making these technical comments, we are aware of the many forces that were acting against such recommendations, set in motion well before 1998. • Moreover, with PMT being dominated by CA membership, and with this membership comprising elected RDC representatives, it was inevitable that the PMT shared projects equally amongst the CA membership, rather than responding to commercial opportunities or implementation performance. Hence the situation where each district was 'entitled' to one large project each; where commercial criteria and innovation were secondary; and where a sustainable mechanism failed to develop. • In large part, the increasing burden on DA resulted from the marginalisation of CCG membership and the loss of capacity by several of these NGOs. The net result was to replace a system that might have become sustainable, with a system that relied on built capacity in an Institutional Contractor that, by definition, was phased out with the Project. An opportunity to build local civic capacity was lost. This was not the Institutional Contractor's intention, although towards the end of the project it was forced to taking on a more assertive role in CDF to complete these projects when it was clear that CA did not have the capacity for this. One can also argue that the increasing burden on the Institutional Contract resulted from the demise of the CCG partners, with CASS and ZimTrust in particular losing capacity. NRMPfolNALREI'ORT 2lJJUI .. Y2003 38 • Moreover, had CCG members been encouraged to use CAMPFIRE to diversify their financial portfolios and funding sources, capacity and support is likely to have extended beyond the withdrawal of USAIO, which is not now the case. A more unified programme, and endorsement of CCG funding proposals by CA, might well have been important in this regard. • Nevertheless, the most important outcome was that the spreading of investment did purchase the political survival of CAMPFIRE, albeit at some cost in terms of technical and economic performance. As will be reinforced throughout this section, the evaluation is not in agreement with the conversion of CCG members into CSPs, as this removes the critical dynamic of long-term relationship-based facilitation and vision from the program. NRMP was correct in trying to infuse performance related grants into the programme, and to use consultants more, but this was based on an assumption that the NGOs that initiated CAMPFIRE would continue to perform and provide quality long-term services and leadership. Unfortunately, this was not the case, and NRMP suffered consequently. In retrospect, the assumption of continued NGO performance and leadership is central to the project design, and should therefore have been an explicit design assumption with a tracking mechanism to ensure that corrective action was taken. The Increasing role of the IC was the implicit response to this problem, but for reasons of sustainability may not have been the best approach in the long-term interests of CAMPFIRE. As we suggest in the lessons learned section, a three-component approach to supporting such programmes should be considered in project design. To incorporate critical long-term leadership and facilitation into CAMPFIRE-type programs: • Provide small, long-term, core, support grants to key staff in key agencies. • Subject this grant to continual performance appraisal, with a strong emphasis on impact at beneficiary level. (tracking performance assumptions would also trigger corrective action). • To link funding to performance, over and above core long-term funding, allow support agencies to grow by competing for additional CSP-type funding on a competitive and performance basis (e.g., a voucher￾type system). 1.4. Recommendation Shift the NRMP's program focus from expanding coverage to graduating communities. Emphasize "getting it right'" in promising areas, or where promising initiatives show the most rapid progress; Refine the criteria for what constitutes a successful CAMPFIRE community that can be considered graduated; CCG implementing organizations should develop planning approaches that include a withdrawal phase, after which outside technical assistance would be intermittent and minimal; Develop a CAMPFIRE approach to CBNRM (and other indigenous resources). In concert with the District Environmental Action Plan (DEAP) and other initiatives, define and generalize a flexible model of participatory land-use planninglCBNRM that is linked to a set of "best practices" guidelines that can be adapted to differing contexts. The CA should sanction this approach and include provision for the "graduation of CAMPFIRE communities" and an "exit strategy" for CCG partners. Evaluation Comments Most of these recommendations pertain to technical issues and the need to refine approaches as the NRMP nears it conclusion. These recommendations were not actively pursued, as stakeholders appeared to believe that the sun would never set under the NRMP. CA's predilection was to spread resources to more ROCs rather than concentrate on a smaller number of more promising initiatives. CA's ROC membership also increased dramatically over the last few years; there are now more hungry ROCs to feed, coupled with a need for CA to produce incentives/benefits to these new members. This trend (diminution of technical considerations in NRMP implementation) was probably exacerbated by the gradual withdrawal of CSPs (who tend to be more technically oriented) from the NRMP decision-making process. "Getting it Right" and Community Graduation As already discussed, the loss of technical capacity and adaptive management processes within the CAMPFIRE movement in the latter stages of NRMP II acted directly against the requirements and hopes of the Mid-Term recommendations. NRMPFINAl.REPORT 29.1ULY2003 39 The concept of "graduation" is erroneous if one believes in rewarding performance, in investing in success, or in utilizing capacity improvements: the better a community performs, the more likely that returns to investment will be high. Nevertheless, USAID responded to this recommendation and developed a criteria which were monitored with ZimTrust assistance and reported annually for several years in formal project reports to Washington, D.C. This stopped when ZT stopped monitoring, and USAID Washington reduced the size of annual reports as documented in USAID Annual Reports to D.C. While USAID-Zimbabwe did it as the Mid Term Evaluation recommended, they also did not agree the concept for the same reasons suggested by this evaluation (Loken, comments). If we take "graduation" to mean sustain ability, the early wildlife communities and hunting enterprises, are, in the majority, sustainable. Several of the ecotourism enterprises might also become sustainable, provided they are managed at a local level and benefits are not spread too thin. The most viable of these were being discussed or developed before NRMP II (e.g. Muzarabani, Kairezi, Chimanimani), but the CDF enhance implementation. Suslainabllily How much of the sustain ability of the wildlife-based programmes is attributable to NRMP II requires discussion: • NRMP II capacity-building obviously helped, especially NRM techniques and procedures. • The counter suggestion is that the maintenance of this capacity at RDC level also entailed higher operating costs at RDC level, and thus reduced benefit flows to communities. However, our data suggests that this was not the case. While there are some inefficiencies at RDC CAMPFIRE Unit level, and some functions are over-centralized, CAMPFIRE units in the main are reasonably managed. The reduction in revenues devolved to communities is not linked to the size of CAMPFIRE Unit budgets, but to the extraction of revenues by RDCs for other purposes. • That sustainability is closely related to the initial CAMPFIRE districts, attests to the initial design and focus of CAMPFIRE on high-value enterprises, when the program had little donor funding, and financial productivity and sustainability were central issues. The early stages of CAMPFIRE were astute, taking advantage of the quickest, easiest and most lucrative opportunities to promote the concept of devolved NRM. NRMP II was implemented under difficult diversification objectives and country conditions. The difficult country conditions, loss of leadership and technical skills, and the acrimony within the CAMPFIRE movement at this time was the root cause of the design flaws in NRMP II, and therefore its sustainability problems. It also took on the far more difficult task of diversification, and while this was the correct decision and is likely to have favourable long-term results, at least if the Zimbabwean tourism economy recovers, too little was done in the way of product development, demand analysis and marketing. Product development is also a process with a much longer￾term profile than relatively short-term donor projects, and this recognition needs to be built into project design. Project versus Programme The structure of the Project, with a heavy funding cycle followed by phase out characteristic of the typical project cycle, is not ideally suited to the programmatic approach of CAMPFIRE with its emphasis on institutional behavioural change and self-sufficiency. The question of the relationship between a long-term programme like CAMPFIRE and a short-term injection of project funding through NRMP deserved more critical analysis and thought in its design. Regional experience suggests that institutional and behavioural change is a process that requires, at a minimum, some 20 to 30 years, especially given the vicissitudes of economic and political stability in emerging democracies. The supplementation of a programmatic approach with a large project, changed the nature of a lightly funded, bottom up and demand-driven program. and increased the emphasis on accessing heavy. increasingly top-down funding: • The relatively light, but high powered, CCG was replaced by a much larger contingent of middle managers and consultants. and a heavier project administrative super-structure. Given that USD3.5 million was disbursed as grants to RDCs and communities, this structure absorbed some USD15 million over seven years. of which at least 20% can be attributed to CDF support (Table 11 ). NRMPFINALREPOR I 2'lJUL Y2003 40 • Even after cutting the initial plans for the COF from US06 million to US03.5 million, a great deal of external effort was needed (in the form of OA and consultants) to spend this money within the time frame of the project. • The efficacy of such a supply-driven approach for the spontaneous development of local capacity is questionable. Without being linked directly to the (often slow) process of building local capacity, the sustainability of individual projects is less likely (if people learn to build things themselves, they are more likely to maintain them). Also questionable, is the likelihood of replication and growth using other forms of funding. • It did not help that the heavy funding phase and the scaling up and diversification coincided with the collapse of the Zimbabwean economy. These approaches also changed the nature of the CAMPFIRE program. While NRMP II may not have been the direct cause of this change, it exacerbated the problem through a combination of more centralised project management structures and large amounts of 'free' money. (CAMPFIRE was envisaged as a process of institutional and behavioural change. It hoped to shift from command-and-control, central planned and open-access systems, to the internalisation of costs and benefits, and management at local level with resource allocatory decisions being made according to comparative advantage. Many of the implementers did not see CAMPFIRE primarily a wildlife programme, but as a process that used the high value of wildlife to catalyse institutional change. Hence, the promulgators were not too concerned about diversification in the belief that once sound institutional structures were developed for wildlife, it was inevitable that these would also manage all natural resources. There was the added inSight that while NGOs and government agencies might think sectorally, communities did not: living with resources, and understanding the interactions and tradeoffs between them, their automatic response was to integrate resource planning and use). NRMP II built the capacity to manage projects. However, being forced to work increasingly through government agencies and councils, it under-emphasized the importance of long-term institutional processes and self-motivated development. It was concerned more with implementing grant-funded projects, than the intricacies and elegance of institutional change. Nevertheless, in expanding the geographical base of CAMFPIRE, NRMP II was important in spreading the application of the original CAMPFIRE prinCiples, especially after it recognised the problem of working mainly through ROCs and started working more directly with communities . That principles, such as benefits to the producer community have survived, is a testament to the rigour and resilience of the original CAMPFIRE principles. Support Model A small amount of high quality field-level technical assistance, spread over a longer time frame, is suited to CAMPFIRE-type institutional change programs. Encapsulated in a statement by Chiredzi ROC is an important lesson. The belief that "we [Chiredzi ROC] can do everything that ZimTrust can do," demonstrates that support agencies, including CA. OA and CSPs, should have placed less emphasis on expanding their middle-management capacity. A better model would have been to build high-level technical capacity centrally (as was initially the case in the formative period of CAMPFIRE), for the purpose of supporting middle managers at ROC level where they are most effective. In this regard, NRMP II support to ROC CAMPFIRE Units, was an valuable contribution. especially in the wildlife districts that could maintain these (we are less sure of this conclusion in ROCs with insufficient funding to sustain the ICB-initiated capacity after donor withdrawal). Where NGOs were able to support these units with good technical capacity (e.g. WWF and NRM technology and training), the model was effective. However, where they effectively replicated ROC capacity (e.g. ZimTrust's "army" of Area and Regional Managers) or were unable to provide support (e.g. CASS) it was not. The central assumptions of service provision (as opposed to long-term relationships) and of CSP withdrawal both need to be questioned. as does the efficacy of the traditional project cycle. What is evident on the ground is that 'graduating' communities, should not have their funding phased out, as they are exactly those communities that have developed the capacity to absorb funding productively. Chiredzi, for instance, has developed the capacity to manage finances and equipment well, the confidence to outsource and work eye-to-eye with the private sector, and now has the ability to plan strategically, and to facilitate communities to understand this vision and move towards it. Instead of 'graduation' and the phasing out NRMPFINALREPORT 29JUL Y2003 41 funding, Chiredzi is now ready to absorb tens of millions of dollars to convert the Lowveld into a major tourism-irrigation economy (see Chiredzi ROC, 2002; de la Harpe, 2002a, b). GGG Phase Out The approach of NRMP II Project affected the sustainability of support to institutional change. The short￾term 'feast' of contracted (rather than partnership) technical support is likely to contribute to a long term 'famine' of relationship-based technical assistance. The blame for this should not be laid directly at the door of NRMP. However, NRMP II did give CA the financial capacity to work independently and reduced the need for CA to build long-term partnerships with NGOs. Moreover, the performance of NRMP II was weakened by the loss of leadership within the CCG, and one asks if more could not have been done to retain such leadership within the larger programme? Certainly, the IC was one means of doing this, albeit with a serious question over long-term sustainability. However, at bottom the NGOs within the CCG also bear much of the blame for this loss of leadership as they proved unable to attract and retain high-level skills, or to negotiate a continued role in CAMPFIRE (albeit in difficult circumstances). The conversion of CCG members into CSPs was driven by the unspoken CA "policy" to cut them out of the projecUprogramme. It was used to soften the blow of the loss of direct funding support and, to the extent possible, to keep the CCG members engaged in project activities. We have emphasised the importance of long-term quality support, and the detrimental effect to the CAMPFIRE programme of losing these skills. Core funding is important for maintaining such staff, and by reducing core funding the programme de facto lost high-level programmatic capacity. However, several of the NGOs were not in any case using core funding effectively in this manner, and were expanding the quantity of middle-managers rather than improving the quality of specialist staff. Yet, the programme really needed this capacity, to provide leadership and, in technical terms, to develop the institutional programmes (e.g. POMS). The challenge of NGO performance. and the associated problem of free money (Le. core funding) and weak performance, is certainly not unique to CAMPFIRE, and can be especially problematic where the correction of past inequalities overshadows meritocracy. NRMPFINALREPORT 29JUL Y2003 42 II. Investment Inflows under NRMP I and II On August 29, 1989, USAID authorized the regional NRMP (690-0251) at USD 19,539,000 to provide assistance to three member countries of the Southern African Development Coordination Conference (then called SADCC and now known as SADC) - Botswana, Zambia and Zimbabwe. Funding was also provided to the Ministry of Forestry and Natural Resources of Malawi in its role as serving as the SADC sector coordinator for Forestry, Fisheries and Wildlife. Funds for the Zimbabwe component of this project (690- 0251.13) were obligated on August 31,1989, and January 11,1990, for a total Phase I obligation of USD 7,600,000. As the NRMP was extended into Phase II, obligated investments in CAMPFIRE totalled USD 28,100,000 and was allocated as summarised in Table 4. TABLE 4: OVERALL BUDGET FOR NRMP NRMP INVESTMENTS INITIALLY PLANNED (in '000 USD) Activity Category NRMPI NRMPII TOTAL LOP Wildlife Conservation (WWF, DNPWLM) 1,277 3,070 4,347 Community Development (ZIMTRUST, CAMPFIRE Assoc.) 3,900 3,915 7,815 Community Funds (Community Projects and RDC develop.) 0 6,000 6,000 Regional Communication and Training (ART) 899 1,200 2,099 Planning and Applied Research (CASS, Action Magazine) 700 2,110 2,810 Administrative Contract (Institutional Contractor) 0 2,675 2,675 USAID Management 0 1,250 1,250 Audit 100 80 180 Evaluation 0 200 200 Contingency 344 344 Inflation 380 380 TOTALS 7,600 20,500 28,100 This budget was funded by USAID-Zimbabwe and by USAID's (then) Southern Africa Regional Programme (SARP). NRMP I was entirely funded from regional funds, and USD4.5 million of NRMP II was also provided by regional funds . The regional contribution to NRMP I & II budgets was as follows (Table 5): TABLE 5: SUMMARY OF THE REGIONAL FUNDING COMPONENT OF NRMP Summary of Regional Component of NRMP FY 89-94 FY 94-99 LOP Actual Variance Wildlife Conservation (WWF, DNPWLM) 1,481,500 1,500,000 2,981,500 3221205 -239,705 Community Development (ZimTrust, 4,180,500 1,950,000 6,130,500 6130500 0 CAMPFIRE Assoc.) Regional Communication and Training 899,295 0 899,295 965622 -66,327 (ART) Planning and Applied Research (CASS, 699,000 1,050,000 1,749,000 1749000 0 Action Magazine) Audit and Evaluation 100,000 100,000 29285 70,715 Contingency 239,705 239,705 239,705 7,600,000 4,500,000 12,100,000 12,095,612 4,388 NRMI'FINALREI'ORT :!lJ,IUI.Y2003 43 This budget was revised after the Mid-Term Evaluation in 1998 and the Project Grant Agreement Amendment Number 6 documents an illustrative financial plan as follows (Table 6): TABLE 6: NRMP" BUDGET AS MODIFIED AFTER MID-TERM EVALUATION NRMP " Jin USD) REVISIONS 1998 CUMULATIVE OBLIGATIONS Activity Category USAID GOZ TOTAL USAID GOZ TOTAL Wildlife Conservation 0 800,000 800,000 4,421,205 4,000,000 8,421,205 Community Development 1,200,000 1,150,000 2,350,000 14,070,500 5,400,000 19,470,500 Regional Communication and Training 0 0 0 1,470,010 0 1,470,010 Planning and Applied Research 0 0 0 2,809,000 0 2,809,000 Administrative Contract (Institutional Contractor) 1,700,000 0 1,700,000 4,100,000 0 4,100,000 USAID Management 0 0 0 1,000,000 0 1,000,000 AudiVEvaluation 0 0 0 229,285 0 229,285 Contingency 0 0 0 0 0 0 Inflation 0 0 0 0 0 0 TOTALS 2,900,000 1,950,000 4,850,000 28,100,000 9,400,000 37,500,000 The contract of the Institutional Contractor was extended in the post 1998 period. This did not increase the total project budget under Phase II as adjustments were made within existing budget lines. Initially under NRMP II, the use of project funds was anticipated to be as follows (Table 7): TABLE 7: BREAKDOWN OF NRMP " BUDGET By PROJECT COMPONENTS AND IMPLEMENTING AGENCY ! Ins! I Natural Operation I Institutions Stren thellResource Long Short Training Commodi sl Audit Evaluatio TOTALS term TA Term TA ties Overhead n en s IC 0 0 1,000 450 25 250 950 0 0 2,675 CDF 3,000 3,000 0 0 0 0 0 0 0 6,000 CA 0 0 0 625 350 435 340 0 0 1,750 ZimTrust 0 0 0 525 340 350 750 0 0 1,965 ART 0 0 0 500 250 150 300 0 0 1,200 Action 0 0 0 200 600 180 80 0 0 1,060 WWF 0 0 0 500 450 120 500 0 0 1,570 CASS 0 0 0 500 200 250 100 0 0 1,050 DNPWLM 0 0 0 500 100 400 500 0 0 1,500 MLGRUD 0 0 0 100 40 40 20 0 0 200 USAID 0 0 500 600 0 50 100 80 200 1,530 TOTALS 3,000 3,000 1,500 4,500 2,355 2,225 3,640 80 200 20,500 NRMI'FINAI.REI'Oln 29JULY200J 44 By the end of June, 2003, USAID-Zimbabwe had disbursed USD 13,958,175 million, with an amount of USD 1,641,556 remaining (Table 8). TABLE 8: USAID-ZIMBABWE NRMP II VARIANCE ANALYSIS Source: Analysis of raw data from USAID commitment records USAID -Zimbabwe NRMP II Variance Analysis COMMITMENT DESCRIPTION Committed Disbursed Administrative Contracts 4,100,000 4,034,451 USAID Management 1,048,707 852,255 Community Development 7,603,889 6,383,039 Wildlife Conservation 1,108,857 1,091,642 Regional Communication and Training 513,277 452,652 Studies and Evaluations 249,959 243,081 Direct Grants (to CSPs) 975,042 901 ,055 TOTAL 15,599,731 13,958,175 Variance 65,549 196,452 1,220,850 17,215 60,625 6,878 73,987 1,641,556 Table 9 (following) is an analysis of the same data extracted from USAID's commitment records for the USAID-Zimbabwe component of NRMP II, but provides a more detailed breakdown of expenditure according to activities . The data is further divided into USAID direct disbursements, and into disbursements through the Institutional Contractor. Almost USD14 million out of a budget of USD15.6 million has been spent. TABLE 9: BREAKDOWN OF USAID-ZIMBABWE DISBURSEMENTS BY ACTIVITY FOR SUB-CONTRACTORS Summary of USAID Disbursements for NRMP II (as at June 2003) COMMITMENT Grantee Administrative Manage- Community Wildlife Regional Comms Studies and Direct TOTAL DESCRtPTtON Contracts ment Development Conservation and Training Evaluations Grants Project Budgel 4,100,000 1,048,707 7.603,889 1,108,857 513,277 249,959 975,042 15,599,731 Disbursement (June 2003) 4,034,451 852,255 6,383,039 1,091,642 452,652 243,081 901,055 13,958,175 Variance 65,549 196,452 1,220,850 17,215 60,625 6,878 73,987 1,641,556 USAID Cosls - 636,065 - - - -. 636,065 Studies - - - 10,018 243,081 - 253,099 Development DA 4.034,451 - 1,913,279 - - - - 5,947,730 Associates CDF - 2,986,962 - - - - 2,986,962 WWF - - - 1,086,156 - - - 1,086,156 CASS - - - - - - 401,055 401,055 Direct Grants CA - 216,190 - 230 16,522 - - 232,942 from USAID ZimTrust - - 1,452,591 - 425,451 - 500,000 2,378,042 DNR - 10,845 - - - - 10,845 FC - - 9.988 - 661 - 10,649 MLG - - 9,374 5,256 - - 14,630 Total Disbursement 4,034,451 852,255 6.383.039 1.091,642 452,652 243,081 901,055 13,958,175 NIUvll'l·INAI.REPOI{ I 2').11 Jt Y2(0) 45 By the end of March 2003, expenditures on obligated contract agreements administered through Development Associates were as follows (Table 10): TABLE 10: SUMMARY OF EXPENDITURE ADMINISTERED BY DEVELOPMENT ASSOCIATES (15 JUNE 2003) Source: Development Associate's Records Amount In USD DACONTRACT Funds allocated Obligated Expenditures Total Remaining Per AIDIIC Contract to Contract for Expenditures Unexpended Agreement March 2003 To date Obligated Amount DAJPW operations 5,351,148.00 5,351,148.00 33,820.73 4,943,522.56 407,625.44 WWF Long-Term Grant 1,551,348.47 1,551,348.47 0.00 1,542,099.27 9,249.20 WWF Short-Term Grants 18,063.80 18,063.80 0.00 0.00 18,063.80 CASS 743,950.23 743,950.23 0.00 743,950.23 0.00 CA 1,567,852.41 1,567,852.41 0.00 1 ,567,852.41 0.00 ZimTrust 943,001 .86 943,001.86 0.00 943,001 .86 0.00 SAFIRE 153,028.09 153,028.09 0.00 153,028.09 0.00 RDCs 3,287,554.47 3,287,554.47 15,754.14 3,198,727.42 88,827.05 Sales tax refunds (17,438.62) 17,438.62 RDC/CDF Grant Allocation 712,445.53 712,445.53 0.00 0.00 712,445.53 Support Grant Allocation 61,281 .14 61,281.14 0.00 0.00 61 ,281.14 Total 14,389,674.00 14,389,674,00 49,574.87 13,074,743.22 1,314,930.78 Interviews with Development Associates suggest that the following money will not be spent: • An amount of USD712,445 remains unexpended on CDF budget line. • Over US 88,827 remains unutilised under the support grant budget line, intended for CSP small grants by applications. Table 11 (following) summarises NRMP II expenditure as at June 20039 . In the absence of a summary of regional expenditure by budget line, these amounts are estimated. Expenditure of USAID-Zimbabwe funding is extracted from DA records as of June 2003. It is difficult to analyse CDF data in more detail because records are maintained in ZD where financial figures are rendered almost meaningless by inflation and devaluation. We have followed DA's conclusion that USD2 million was utilised on Institutional Capacity Building, with the remaining USD2 million allocated to the CDF of which approximately USD750,000 is not expected to be used. Notwithstanding these minor uncertainties In the financial data, Table 11 provides a valuable overview of the project. 9 Note that compilation of this data has been difficult. USAID keeps a full record of commitments, but does not as a matter of course keep management accounts that list the three critical criteria: budget line; agency; date. Furthermore, the inclusion of both USAID-Zimbabwe and regional funding has further complicated the collation and summary of data. This is indicative of the absence ofa consol idated Management Information System forNRMP II. NIUvll'l·IN:\I.REI'C)RT 46 TABLE 11: OVERALL SUMMARY OF NRMP II EXPENDITURE Summary NRMP II Expenditure Budget Expenditure Percent Agency USAID/Zim Regional Actual Variance Of Actual WWFIDNPWLM 1,500,000 1,500,000 - 9% ZimTrusUCA 1,950,000 1,950,000 - 11% CASS/Action 1,050,000 1,050,000 - 6% DA 5,351,148 4,943,523 407,625 28% WWF L T Grants 1,551,348 1,542,099 9,249 9% WWF ST Grants 18,064 - 18,064 0% CASS Grant 743,950 743,950 - 4% CA Grant 1,567,852 1,567,852 - 9% ZimTrust Grant 943,002 943,002 - 5% SAFIRE Grant 153,028 153,028 - 1% Other Grants 61,281 - 61,281 0% Sales Tax Refund (17,439) 17,439 0% ROC ICB (approx.) 2,000,000 2,000,000 - 11% CDF 2,000,000 1,198,727 801,273 7% 14,389,674 4,500,000 17,574,743 1,314,931 Table 12 allocates the revenues presented above in the form of an input-output analysis. While percentages are unavoidably indicative, this nevertheless is useful for assessing the effectiveness of various project components. We have been criticised by both DA and USAID about these figures, and while we accept that they are indicative we believe that an input output analysis adds value to the evaluation and is helpful in showing where money worked best, and where it could have worked beUer. Both USAID and DA accept that the project should have monitored financial data against outputs in the first place to allow a more accurate appraisal. We take at face value DA's statement that roughly half of the DA and ICB grant was targeted towards implementing CDF projects, and roughly half was for other capacity-building purposes 10. However, as noted below, surely a significant amount of general capacity building must also have been aimed at the implementation of CDF, for why else invest in the non-wildlife districts (which comprise 16 out of 29 Large CDF Grants)? to DA (comments) suggests that thc mitial allocation of 40% to CDF support by the evaluation was incorrect. While no empirical data IS available, DA suggests that a better approximation is 15-20%. This assumes that 35-45% of the amount spent on ICB and 35-45% of DA expenditure can be attributed to supporting the CDF. DA also notes that ICB grants were used not only to support the CDF but to support I\RM and other aCllvities II1cludmg "World Bank Capacity Building Programme for RDCs, various UN sponsoreci povcrty reliefpr0gnlms, speCial NGO funded projccts, DEAP, etc. -most of these activities were never intended to result In a CDF mfrastructure project" DA "ls0 notes" As per the DA contract a great deal of time/resources wcre spent on financial and coml1locilly management trall1ing- much of which was not solely related to cor infrastructure projects". However, DA docs agree that 111 the last two or three years it might well have secmcd that their activitics were focused almost entirely on the CDF. We also accept DA's statement "A Iso, there were considerable traming and T A efforts made for CA - various strategic planning exercises, personnel manuals, endowment fund TA, stafftraming systems, Board training, financial mgt training. Much time was spent on organl7.lng (with CA quarterly Programme reviews, coordination mcetings with CSPs, etc, etc." Unfortunately, due to the absence of a comprehensive management information system, and as we have already noted in the preamble to this argument, it was not possible to draw out accurate data to compile the input:output analysis, which therefore remall1s indieatlvc. Although we wonder what the objectives of the ICB were in (particularly) the non-wildlife districts ifit was not to creatc awareness with a CDF project in mind, we also accept DA's argument that attributing the full 40% to CDF support is probably an exaggeratIOn. However, this docs not change the basic conclusion that CDF delivery had lower returns to investment than other components of the programme, largely because of design flaws and political and personality considerations that prevented these being corrected (see section IV.A). NRMI'FINALREPORT 29JUL Y2003 47 TABLE 12: ROUGH INPUT-OUTPUT ASSESSMENT OF NRMPII Output Improved natural resource management and revenues General institutional capacity building General capacity building (CA, RDCs) Capacity-building targeted specifically at CDF CDF Projects Regional communication Input 18% 25% 20% 20% 7% 10% Natural Resource Management: The main contributions to improved natural resource management were: • The design of the CAMPFIRE programme, with wildlife revenues of approximately USD2 million annually. • Approximately 18% of NRMP II revenues for wildlife management, which passed largely through WWF. • Improvements in general institutional building, for instance the development of CAMPFIRE Units at ROC level by the ICB. The output/input ratio of this component of the project has been high as illustrated byt eh impact of the programme (Section III A and B). Instilutional Development: The main strengths of institutional development were the sound principles embodied in the CAMPFIRE plus the wildlife revenues flowing through these producer communities and RDCs. This has led to generally sustainable and well-governed community institutions. Unfortunately, opportunities to build on this platform were not taken. For example, institutional capacity has not been well monitored by the NRMP II so this progress cannot be quantified. Opportunities to use the CAMPFIRE experience to develop improved systems and training materials for institutional management have not been used. The support of CAMPFIRE Units, as noted above, has been helpful, although a proportion of this was related directly to helping RDCs cope with financial and commodity matters related to USAID regulations and requirements. We assume that the majority of ZimTrust and CA expenditure (25% of budget) and some of the DAIICB costs (10-20% of budget) were allocated for this purpose. The effectiveness of this component of the project has been moderate, and there is certainly considerable scope for improvement in delivery. NRMP II lacked an institution that was highly skilled in institutional development. The 20% of the NRMP II budget spent variously on capacity building is too aggregated to be able to comment specifically on its impact, except to say that CA was not a good recipient of such help. Community Development Fund: The CDF used considerable project time and resources, especially in the latter stages of project. This component used approximately 20% of the budget (i.e. half of the DA plus ICB grants) and delivered projects to the value of 7% of project budget. As noted elsewhere, the design of this component was unsophisticated, implementation was slow and cumbersome, and the original CCG was not committed, ready, nor tooled to administer or support grant funding. We have already commented on the dangers of 'free money'. The project's later emphasis on Trusts and community-level implementation recognised that the initial top-down ROC-level grant funding was problematiC, and not in keeping with the objectives of the CAMPFIRE programme. For these reasons, and the coincidence of this phase with macro-economic decline, projects choice and implementation was sub-optimal from a business perspective (although, it must be said, a good effort was made by DA to complete these projects before closeout). This component of the project had the lowest economic return on investment, but was crucial politically. The project: • Was initially designed to deliver USD6 million to communities in the form of CDF grants. • Used USD2 million for Institutional Capacity Building largely at ROC level (23 grants). • Delivered approximately USD1.3 million as CDF Large and Small Projects. • Rather it aimed to Implemented one Large Project in each participating ROC, with the criteria that they could at least be maintained (it did not use return of investment or profitability as a criteria, and with grant funding there was no requirement to cover the cost of capital). Project resource allocations and design criteria were not rigorously applied throughout implementation due to such NIUvlPI'INA LRI,POR I 2'JJlII .Y200J 48 extraneous considerations as equity, community desires/preferences and natural resource management needs (Loken, comments). • CDF project investment comprises 7% of total project expenditure, with a significant portion of this investment including consulting fees for feasibility studies. • More than USD $700,000 remained unspent. • A further USD 2m was used to capacitate RDCs to deliver these projects (11 % of NRMP II) but RDC/ICB grants designed to accomplish much more than just CDF-related capacity such as district-level NRM, community awareness and capacity building. Regional and International Communication: The investment in regional and international communication, largely through ART, CASS and ACTION, has had a powerful impact, through publications and at international conservation forums. Indeed, the funding provided by USAID in combination with the powerful philosophy of the CAMPFIRE movement, has had a significant impact on the direction of international conservation, and in particular the growth of interest in sustainable development and CBNRM. People as far away as US education institutions, Mongolian wildlife managers and Asian foresters and aquatic managers refer to CAMPFIRE. For reasons related to pressure from the US, funding to Africa Resources Trust (which was central to the impact of regional and international communication and Zimbabwe's success at CITES in 1997) was discontinued. Consequently, this component lost effectiveness, although USAID tried to provide some continuation by providing direct support for attendance at meetings. It must also be said that both Action Magazine and CASS lost the capacity to contribute effectively, and consequently did not have their NRMP II funding continued. In summary: • The returns on Investment in wildlife management and regional and international communication have been high. This is because the institutions managing these components were technically strong, and they built on the strengths of CAMPFIRE to present a powerful message internationally and locally. • The returns on investment in capacity building were mixed, good for some CAMPFIRE Units, poor in others and in CA, and a major opportunity to develop lower level institutional mechanisms was lost. Outside of NRM and advocacy, this is because community capacity building skills within the programme were lacking resulting in mediocre implementation. • The returns on investment from the CDF were political sustain ability and the spreading of the CAMPFIRE concept widely across the country, with a trade-off in terms of technical and economic criteria and sustain ability where delivery was unexceptional. The value of political sustainability must not be under rated. Difficult political considerations, and the loss of technical skills, that meant that this component was compromised technically and lacked the institutional sophistication associated with the CAMPFIRE programme. The CDF concept was the right thing to do and was useful but in better circumstances could have yielded better results, using more experimentation and piloting, adaptive management and innovation to build on the learning opportunities that CAMPFIRE offers. NRMI'FINALREI'ORT 29JUL Y2003 49 III. Impact Achieved and Challenges Faced With the previous chapter having documented USAIO-funded contributions to CAMPFIRE to date, this chapter assesses what USAIO has "bought" with its support, and the results and impacts, particularly since the 1998 Mid-Term Evaluation A. CONSERVATION 1. Awareness The CAMPFIRE programme is widely known as the "African solution to the African problem" because a lot of awareness activities have been done to make the programme acceptable and justifiable socially, economically, environmentally and politically. Under NRMP II there has been significant investment in the awareness raising activities of ROCs, CAMPFIRE Service Providers and the CAMPFIRE Association itself. Conservation awareness has been promoted through the formation and establishment of Natural Resources Management Committees from village to district level; training workshops targeted at producer communities, conservation awareness among schools, and information dissemination through the print and electronic media. Natural Resources Monitoring: In the post 1997 period, many CAMPFIRE districts recruited Natural Resources Monitors and CAMPFIRE promoters. Table 13 indicates the numbers of monitors and promoters who were recruited by selected ROCs and deployed to monitor and report the state of specific natural resources at village and ward levels. CAMPFIRE promoters function as extension personnel and provide back-up support to ROC training officers. Anecdotal information suggests that monitors and promoters have played an important role in reducing incidents involving un sanctioned harvesting of big and small game, fish, fruits and commercial timber. Natural resources monitors and promoters motivate local communities to take a leading role in the management and sustainable utilization of natural resources. TABLE 13: NATURAL RESOURCES MONITORS AND CAMPFIRE PROMOTERS RECRUITED AFTER 1997 Rural District Council Natural Resource CAMPFIRE Monitors Promoters Nkayi 6 - Mudzi 6 3 RushinQa 10 7 Nyaminyami 11 - Tsholotsho 8 - Uzumba/Maramba/Pfunowe 10 4 Beitbridge 12 1 Guruve 14 - Muzarabani 10 - Bulilimamangwe 10 - Gwanda 5 5 Chlmanimani 4 - NyanQa 4 4 Blnoa 7 - Kusile (Lupane) 6 - North 6 - 129 24 Information dissemination: Action Magazine has played a critical role in disseminating conservation awareness messages through schools and teachers training colleges countrywide. Action has produced 12 issues of its magazine, with an estimated circulation of 200,000 per issue covering six countries in Southern Africa. The CAMPFIRE Teachers' Pages has been distributed to 100,000 teachers and about 15,000 schools, 6,597 of which are in Zimbabwe. Action estimates that this environmental awareness drive has reached 1.5 million school children. In addition to magazines, Action's products include: NRMPFINALREPORT 29JULY2003 50 • An ecosystem poster set, which comprises a series of four posters featuring wetlands, grasslands, deserts and woodlands. • Animal poster set - this comprises a series of 13 full colour posters of wildlife. These are often found in CAMPFIRE districts. • Action's training, curriculum and research unit, in conjunction with the Ministry of Education, Sport and Culture, is spearheading a dynamic and participatory approach to the teaching of Environmental Education in Zimbabwe. The CAMPFIRE Association on its part has contributed towards conservation awareness by: • Participating at major exhibitions especially the Environmental Expo; • Participating at the Zimbabwe Agricultural Show; • Placing editorial supplements in Zimbabwe's major news daily newspapers WWF has also contributed to conservation awareness. The CAMPFIRE Game has been incorporated into school curricula. Posters are widespread. especially one on managing and counting elephants. Awareness and training: Through training provided by WWF and ROC officials, producer communities are encouraged to undertake their own wildlife censuses. They later hold meetings with Parks Officers to compare their census results and determine sustainable off-take of wildlife. The Impact of all the activities associated with conservation awareness is reflected in the efforts being made in protecting landscapes and wildlife habitats, and in the interest shown by communities through the small and large projects (see section "I.C on diversification). CAMPFIRE has developed income-generating enterprises based on natural resources and these projects are linked to natural resource management strategies. Community actions and attitudes towards these resources have significantly changed. Strategies that have been embarked on by communities include: formulation of by-laws that govern access to the resources, fencing the resource, monitoring and protecting resources, establishment of village natural resource management committees, conducting Environmental Impact Assessments and periodic resource audits. Perhaps the primary awareness mechanism has been the participation of tens of thousands of households each year in deciding how to allocate wildlife revenues. Most people in CAMPFIRE producer communities (several hundred thousand) are well aware of the programme, and this is confirmed by the Process Orientated Monitoring System (POMS) (ZimTrust 2001,2003). Most conservationists in the country know about it, and random questioning of ordinary people (e.g., border officials) suggests knowledge about CAMPFIRE is widespread. The report by Chikuruo et al (2003) also confirms that local government officials at district and provincial offices are well are of the programme. A number of factors indicate that communities were increasingly aware of the value of wildlife and how to manage it 11. Indicators include: • The number of Councils applying for and being awarded appropriate authority. • The improvement in marketing of wildlife (c.f. Child, 1995a). • The take-over of quota setting by communities and the maturity with which this was done. • The monitoring of safari hunting. • The high level of participation in revenue distribution. • The uptake of problem animal reporting. • The setting aside of dedicated wildlife areas, and the greater respect given to natural resource conservation in CAMPFIRE, as seen in fire control, reduced tree-cutting, reduced poaching, etc. The growing demand by communities to develop ecotourism operations (Tables 23-25) indicates a general awareness of the value of tourism, although this has not yet translated into specific knowledge related to managing such operations. II See table on page 19 of Child et al (1997) NRMI'FINALREI'OR r 29JULY200J 51 2. Protecting Landscapes and Habitats No accurate maps or lists of the area formally or informally set aside for wildlife by communities is available. The primary wildlife producing wards are illustrated in Figure 6. FIGURE 6: MAP OF CAMPFIRE "PRODUCER WARDS" IN 1996 Source: WWF-SARPO database Wards with CAMPFIRE Revenue: 1996 IiURUNGWE • CAMrflRf W.rd Source: WWF Database E KM o 100 200 Communal Areas, especially those in the Sebungwe and Middle-Zambezi Valley have been experiencing population growth rates in excess of 10% annually. This is due in large part to in-migration from overcrowded and degraded areas elsewhere in Zimbabwe (e.g. Save Valley). Within two decades, this 10% human population growth rate transformed much of the Sebungwe and Mid-Zambezi Valley from wilderness into subsistence agriculture of questionable sustainability12, with survival being largely dependent on food relief and other government inputs (CASS, 2000). Bond (2002) demonstrated that what enabled people to live in these marginal areas was heavy subsidisation of many aspects of livelihoods (government food handouts, schooling, health, agricultural inputs, etc.), including the extraction of open access resources. This was especially prominent when new people moved into areas, usually already settled (Derman, 1990). In the face of this, CAMPFIRE protects an area of land roughly equivalent in size to Zimbabwe's Parks and Wildlife Estate. This is a major indicator of the successful impact of the CAMPFIRE programme. The total area of CAMPFIRE Producer Wards is 56,135km2 (see Table 14). Zimbabwe's protected areas, which comprise 12.7% of the country, total 49,700km21J. However, there are 152,255 households (605,691 people) living in CAMPFIRE Wards. Assuming each household requires an average of 5 hectares, the total area protected by CAMPFIRE is 48,522 km2, still roughly equal in size to the Parks and Wildlife Estate. 12 Research by Cumming and Lynam (pers. comms.) quantified these massive increases in human numbers, loss of habitat and unsustainability of land use practices. 1.1 List given on p35 of Child G (1995) NRMPFINIILREI'ORT 2'>JUI .Y200) 52 This area contains roughly 12,000 elephants and 14,000 buffalo, and is under stocked, although wildlife populations are increasing (see section III.A.3). TABLE 14: CAMPFIRE BASELINE DATA Source. WWF database (see Annex J for more information) District Total Area Year Year Total District Natural Granted CAMPFIRE Number Population Region AA Started of Wards Wards arsons) Beltbridge 1991 1991 21 6 80.946 Binga 3,4&5 7,930 27 21 87,802 Bubl 5,547 88 12 ',( 2 36.614 Bulillmamangwe 4&5 12,574 1,530 33 10 156,641 Chaminuka 2a,2b,3,4&5 2,752 380 26 2 94,047 Chimanimani 1,2a,3,4&5 3,419 1994 No activity 28 110,836 Chiplnge{Gazaland) 1,2a,3,4&5 5,223 1991 408 33 2 336,693 Chiredzi(Gaza Khomanani) 5 17,748 1991 1991/92 3,633 32 9 Chiweshe{fy'lazowe) 21;1 4,482 1991 375 29 5 Gokwe North 3,4&5 7,359 1991 1991 2,523 25 4 164,558 Gokwe South 3&4 11,138 1991 . 1,308 28 6 238.581; Goromonzi 2a 2,504 26 147,126 Guruve 2a,3&4 7,810 1988 4,215 28 14 Gwanda 4&5 10,792 1994 No activity 2,283 23 6 112,984 Hurungwe 2a,3,4&5 19,895 1992 1992 2,793 40 9 246,902 Hwange 4&5 29,934 1989 1991 4,021 27 15 71,707 Hwedza 2b&3 998 69,981 Kusile(Lupane) 3&4 7,780 1996 2,885 24 11 94,469 Marondera 2a&2b 3,554 24 104,601 Matobo 4&5 7,278 1,233 26 4 Mudzi 4,222 1992 1 1,009 18 Mutoko 2b.3&4 4,052 29 Muzarabani 2a,3&4 4,322 1991 2,540 17 Mwenezi 4&5 12,933 31 Nkayi 5,333 993 No actMty 2,628 23 Nyaminyami(Karibal 4&5 6,327 1988 1989 3,532 16 11 27,717 \,981 1'\0 1.ab,3&4 5,738 253 37 Pfura(Mt. Darwin) 2a,2b,3&4 1,771 164,362 Rushlnga 2,408 1992 1993 17 Tsholotsho 4 7,823 1990 1989 5,354 20 8 111,828 UMP Zvatalda 2b,3&4 2,682 1993 9 3 619 15 2 86,302 Umzingwane 4 1,074 1988 62,954 Zlva we 4 2363 1998 65752 TOTAl/AVERAGE 247,078 56,135 735 . 165 4,000,710 NRMPFINAI.RFPOI{ r 2 Note that these data are averages of the averages of each concession area. A detailed analysis is required that includes number in sample and standard error, but is beyond the scope of this evaluation required. NRMPFINALREPORT 29JULY2003 58 B. ECONOMIC BENEFIT Economically, CAMPFIRE can be assessed in three stages. It was initially based on the sale of high value safari hunting. In advantageous locations, high-end tourism lodges were then developed. Finally, and funded largely by NRMP II, investments in ecotourism (usually small tourism ventures within communal lands targeted at specialised - ornithology, culture - or middle to lower market segments - camping and chalets) were made. NRMP II also funded a variety of projects aimed at developing other natural resource (NR) products, too (e.g., ilala palm, fish, honey, etc.), largely through the small grants programme. a) Direct Income to CAMPFIRE Areas From 1989, the income from safari hunting increased rapidly to USD 2 million annually. In terms of Zimbabwe's economy, the multiplier on trophy fees is approximately 3, suggesting that CAMPFIRE areas generated direct income from hunting of USD6.0 million to Zimbabwe annually. During the period under review, CAMPFIRE areas earned USD 20.3 million, and Muir-Leresche (2003) estimates that this earned over USD 100 million for Zimbabwe. In addition to direct fees to communities, a proportion of safari operating expenses is paid locally as wages and salaries (some USDO.S million annually). and for the purchase of materials. but this is not captured in CAMPFIRE monitoring records . Neither are economic multiplies like taxidermy. travel. extended tourism activities, food, etc. FIGURE 13: DIRECT INCOME TO CAMPFIRE Source: WWF database; see Annex L for further details -_._------------ -------------- CAMPFIRE: Sources of Direct Revenue (US$) 53.000.000 52.500.000 52.000.000 51.500.000 SI.ooo.OOO S500.ooo iOOther(4) o Hides & Ivory(3) . Tourism(2) m Sport Hunting( 1) 1989 1m 1991 1992 1993 1994 1995 1998 1997 1998 1999 2000 2001 FIGURE 14: FACTORS CONTRIBUTING TO INCREASING INCOME: IMPROVING MARKETING AND INCREASING WILDLIFE POPULATIONS CAMPFIRE Marketing Efficiency Index of Value of Hunting Quotas 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 ... cfJ'" .... cfJ"" .... cfJ"> .... r# ... cfJ'" ... # ... # ... 0,,<1> .... #' ""r#' .... Source: WWF Databases. [Note that the buffalo quotas 1993-1997. which lead to the false hump In the index value data, need to be checked in the nrimarv data. a task bevond the scone of this evaluationl NRMPFINALREPORT 29JUL Y200) 59 The extremely rapid growth until 1994 can be attributed to two factors (Figure 14): • A steady increase in quotas as, initially, more areas were designated for CAMPFIRE and, later, wildlife populations began to grow; • A significant improvement in marketing 17 While the data indicates that CAMPFIRE continued to market its hunting well, in the last three years the parallel exchange rate played havoc with community incomes and is a serious threat to the programme. In 2001 and 2002, safari operators generally paid for hunting at the official exchange rate of 55, compared to the parallel rate of 500 and 1,500. Communities' income was therefore less than 10% of what it should have been. In many areas, this situation continued into 2003. Some ROCs have begun to adjust, but are still getting less than a third of the real value of their hunting. For example, Guruve now gets Z0473 and Chiredzi Z0800 to the USO, one sixth and one third of the real value of wildlife respectively. The Operators, too, are under pressure, as they are also forced to trade at the official exchange rate. However, the likelihood is that they are able to gain some relief by under-declaring revenues. There is nothing so destructive of economic growth, or regular management, as the need to manage finances in these circumstances. The Project and CA have generally been unable to provide advice to ROCs, or to lobby for more favourable circumstances in the appropriate government agencies. An important factor is that the programme has lost its personnel with economic skills, and has failed to use the private sector to access similar skills. For example, some private landholders are dealing with exactly the same problems (e.g. parallel exchange rates for hunting) and the programme could have learned from them In the early to late 1990s, considerable effort was made by ROCs and CA to capture the benefits from sale of other natural resources such as sand, timber, and kapenta. However, as we explain below, this diversification is not captured by the data, and remains largely unrecognized even after it was pointed out by PWC (2002). Therefore, the second phase of CAMPFIRE expansion is generally associated with a move into high-end tourism. The consultants listed know tourism lodges (Table 18). While generating less money to councils and communities through fees, tourism is important to the Zimbabwe economy and creates significant local jobs and wages. Much of this revenue, including direct fees to ROCs, is not recorded directly as CAMPFIRE income, at least in part because ROCs do not want to share it with communities. Moreover, some lease agreements are weak. For these reasons, the contribution of tourism is under-estimated in the data collected by WWF, as is the contribution of sand, fish, timber, etc. The gaps in Table 18 show that NRMP II has not monitored this aspect of diversification, a responsibility that should have been taken up by WWF. TABLE 18: SUMMARY OF HIGH END TOURISM JOINT VENTURES IN COMMUNAL AREAS Summary of High End Tourism Joint Ventures in Communal Areas District Name of Lodge Beds Direct Income to Jobs Income to Communities Zimbabwe Tsholotsho Ngami HwanQe GorQes 16 BinQa Chizarira 16 Masuma 24 (onlv just open) Nyaminyami Bumi Hills 40+ (closed) KiplinQs 16-20 (closed) Katete LodQe 30 (closed) MsanQu 12 TiQer Bav 40+ Gache Gache (closed) HurunQwe Chioembere (closed) Chipinge Chilo 36 30 Mahenye 16(self catering) Guruve Zambezi LodQe ?? 17 The DNPWLM data collated in the early stages of the CAMPFIRE programme differs significantly from the WWF CAMPFIRE data for this period. Given that the DNPWLM data was based on regular field audits, and on quotas set by DNPWLM, and that WWF was not directly involved in data collection at the time, some doubt must be cast on this early data in the WWF database. While there is minor variation in income data, trophy data and standard values differ significantly in the period 1989 to 1993. NRMPFINALREI'ORT 29JUL Y2003 60 b) Economic Beneficiaries Table 19 shows that CAMPFIRE has been measured to deliver US$20,3 million in direct income for RDCs and communities between 1989 and 2001. This derives from wildlife in some 90-110 Wards, and consistently benefits over 80.000 households (560,000 people). While some districts have stuck well to the concept of producer communities (e.g. Guruve, Chipinge, Chiredzi), in others the dilution of benefits has been problematic. In Sidinda Ward in Hwange, for instance, the majority of benefits were initially returned to Sidinda ward. According to the safari operator, poaching ceased within six months and over the next few years the populations of buffalo and sable increased steadily. However. with one ward outvoted by the non-producing wards, Hwange ROC then began to spread these benefits over some twelve wards. with the ultimate effect that benefits in Sidinda fell, and poaching again increased. Wildlife populations are now at or below their initial levels. The same problems have occurred in other districts (e.g. Tsholotsho) and in other ways, with Nyaminyami, for instance, seriously diluting benefits by appropriating most hunting, tourism and kapenta revenues to the ROC rather then producer communities. TABLE 19: SUMMARY OF CAMPFIRE REVENUES AND WARD AND HOUSEHOLD BENEFICIARIES CAMPFIRE Revenues and Beneficiaries Year Total Income Number of Number of Number of USD Districts Wards Households 1989 348,811 3 15 7,861 1990 556,433 9 41 22,084 1991 776,021 11 57 52,456 1992 1,216,678 12 74 70,311 1993 1,483,873 12 98 90,475 1994 1,642,671 14 101 96,437 1995 1,591,567 14 111 98,964 1996 1,755,912 19 96 85,543 1997 1,837,438 17 98 93,605 1998 1,891,766 15 92 80,498 1999 2,753,958 16 112 95,726 2000 2,105,204 14 108 88,072 2001 2,328,452 14 94 76,683 20,288,784 Source: WWF Database Muir-Leresche et at (2003) calculates that the direct economic impact on Zimbabwe was USD1 OOm. However, PWC (2001) suggests that these revenues under-estimate CAMPFIRE income by at least 30%, as RDCs tend not to account for natural resource income under the CAMPFIRE budget line. Moreover, tourism income is under-estimated as it is seldom recorded and the multiplier on direct revenues is normally at least 10. We can thus conclude that CAMPFIRE's impact on national income is at least USD 10m annually, and may well be higher. If we include the multiplier on tourism activities which, given the high proportion of locally manufactured inputs, has been significant and probably of the order of at least two 18, CAMPFIRE is worth USD20-25m to Zimbabwe's economic income each year. c) Ecollomic Devolutioll In CAMPFIRE's initial years, great prominence was given to the importance of devolution. DNPWLM 'audited' all ROC CAMPFIRE revenues regularly. The importance of producer communities and devolution to the sustainability of income generation was continually emphasized. Performance in terms of fiscal devolution was regularly presented at CCG and CA meetings, and peer review was an important factor in driving devolution, with many Council Secretaries appearing to support this process. The early results (Figure 15) were remarkable, reducing the amount of revenue retained by council from 42% to 13%, increasing the amount reaching communities directly from 54% to 69%, and increasing the amount of 18 It is quoted to exceed 6 in South Africa (Feamhead, personal communications). NRMPFINALREPORT 29JUL Y2003 61 discretionary community income from 20% to 62%. This was not linked to funding with some well-funded districts performing badly (e.g. Nyaminyami (by E.U.) and Hwange (USAID)), and some non-funded districts doing especially well (e.g. Beitbrldge, Chipinge, Guruve). The amount and proportion of revenues devolved to communities is summarised in Figures 1'6 and 17 respectively. Until 1996, fiscal devolution increased steadily until it reached 60% and almost US$1 million, after which no progress was made, the exception being 1999 which included the revenues from the sale of ivory. Indeed, after the millennium, RDCs were extracting around 40% of all CAMPFIRE revenues from communities, with communities getting less than 50% after 1996 and only 38% in 2001. Clearly, early progress was being reversed. This reduced the level of benefits to project beneficiaries. Its also impinged directly on the sustainability of institutional capacity, with the POMS surveys confirming that only those communities that were benefiting directly from CAMPFIRE were willing or able to maintain their institutions. This may also have been a proprietorial issue, as institutions were maintained despite levels of income being relatively low when measured on a per household basis (Figure 18). FIGURE 15: SUMMARY OF EARLY PROGRESS IN FISCAL DEVOLUTION 19 . -----------.----.--------r---,,---c:----, o Council Summary of Early Progress in Devolution o Management • Projects (in communities) 1 00% CJ Devolved 80% 60% 40% 20% 1990 1991 1992 1993 FIGURE 16: SUMMARY OF ALLOCATION OF CAMPFIRE REVENUES TO RDCs AND PRODUCER COMMUNITIES (USD) (1989-2001) CAMPFIRE Allocation (USS): 1989 ·2001 Sl.ooO 12.500 S2.000 Sl.500 51 .000 5500 --+- 50 19 Source: Child et al (1997), which notes exclusion of Chiredzi and Nyarninyarni (because they refused to follow CAMPFIRE principles at this time), and small discrepancies in figures due to difficulties of incorporating carry-overs and different financial years in the accounts from ten districts. NRMPFINALREPORT 29JULY2003 62 FIGURE 17: SUMMARY OF ALLOCATION OF CAMPFIRE REVENUES TO ROCs AND PRODUCER COMMUNITIES (1989-2001 ) Allocation of CAMPFIRE Revenues (%) [J Council (unallocated) [J Council (other) [J Council (Levy) [J m Comrunities 100% 90% 80% 70% 60% 50% 40% 30% 20% 1 0% 0% 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 FIGURE 18: SUMMARY OF AVERAGE HOUSEHOLD BENEFITS FROM CAMFPIRE Source: WWF Database CAMPFIRE Revenue: 1989 - 2001 Actual Benefit per Household (US$) $40 - Vi (/) $30 2- '0 -0 s= - Q) :1 ;. i accountability. In its relationship with the CA, this put the IC in a v ery w~'ward position" 0 tI 0 ! I,' \'J which to administer the grant, and on the other to pl ay the role of mentor, friend, adviser, hEII1lI-llCIldei" ~1I1(J ~J d(lt~ micr clashed with its w atchdog role of guarding financhl Dcc:ounla/)llil'l and, al times . having 10 o manage petty fin ancial m atters. While the IC played the role of policem an well, it was the mentoring and capacity building role that suffered most. Under its contractual obligations, the IC was expected to assist with short term T A in areas of: • Grant proposal development, and/or • Supporting technical analyses (e.g. economic/financial assessments, social limited scope environmental assessments, soundness assessments, and institutional assessments ). b) PriceWaterhouseCoopers DA contracted PriceWaterhouseCoopers (PWC) to provide additional capacity building. PWC supplied March interim Financial Managers (FM) for DA for contract initiation and a permanent Financial Manager from 1997 through September 2002. PWC: • Provided DA with secretarial and bookkeeping services from Contract inception to September 2000. • Conducted pre-award surveys of WWF, CASS and CA and an overhead rate study of WWF in the first few months of the DA Contract, and a second pre-award survey of CA in 1997 followed by its certification as entity eligible to receive USAID funding . Provided extensive, financial management training and capacity building to the CA in preparation for it receiving USAID funds. CA financial certification was acquired in 1997. • Provided advisory services to CA for the development of its long-term grant proposal in mid-1996 (CA only had a bridge-funding arrangement during its first year of operations). • Conducted StrategiC Planning and training exercises for CA starting with a CA Board workshop in February 1996 followed by Technical Assistance (TA) for development of a CA five-year strategic plan and a Board planning workshop held in Nyanga late 1997. • Provided TA for the development of a CAMPFIRE-CA endowmenUfund. • Completed organizational development for CA Uob descriptions, job grading, determination of pay levels, etc.) by October 1997 and followed up with related services in 1998 • Determined the hardware/software and training requirements needed by RDCs to enhance communications with CAlother stakeholders via the introduction of email usage; completed 1999/2000. • Provided TA and training services to improve CASS's financial management of its Grant, completed in 2000. • Concluded its assessment of the CAMPFIRE revenue performance in 2001, and, in mid-2002, presented its findings and recommendations to RDCs in a workshop organized by CA. c) Institutional Capacity Building and the CAMPFIRE Development FUlld Early in phase II, CA and the IC decided to concentrate on implementing institutional capacity building (ICB) projects before embarking on funding community projects. We question the soundness of separating ICBs and CDF projects, and why more use was not made of pilot projects, especially a small grant window, to adaptively develop funding criteria. We also note that there were communication difficulties within the project as many of the initial 150 projects submitted did not fit funding criteria and were rejected. Much of the first three years was spend developing ICB projects (table 29). The first COF project was the Chiredzi Veldt Fire Management signed in October 1999, almost four years after NRMP started, and the second was the Vhimba Eco-Tourism Project signed in August 2000 (almost five years). As table 29 illustrates, the average gap between ICB and COF was about three years. There were 30 ICBs and 21 Large COFs. It took five years to come up with the idea of a small grants window that enabled the communities to actually access the COF, and this then worked well. The first project approved was for Umzingwane on November 6,2000. were Training programmes on the accessing of COF funds, project proposal development and financial matters provided by the IC (often through a sub-contract with PWC) to ROCs. This entailed large, complicated and not very user-friendly manuals. Part of the delay in implementation was caused by the slow development of the COF manual, which was "too complex for communities". This manual is far more NRMPFINALREI'ORT 2'lJUI.Y2003 108 complex than that of other donors (e.g. CIRAO), and the delays in its development is attributed to USAIO imposing a number of complex requirements and becoming over-proprietorial of the manual. Following training, it still took ROCs the better part of a year or more after the initial concept appro'val to come up with detailed project designs that could comply with the stringent requirements of USAIO grant funding under the COF. The absence of a COF manual in the early years of phase II made it all the more difficult for the intended recipients to access COF funds. The COF manual came out late In 1998 and orientation and familiarization of the ROCs and recipient communities with the COF manual seems to have been insufficient. The situation was compounded by a high turnover of staff at ROC level. This meant that those who had been trained may have left their jobs or been transferred to other jurisdictions thus leaving the ROCs without trained personnel to access the COF. Furthermore, within DA there was a high staff turnover, with four different people in the position of Project's Manager within six years. A reluctance to visit field sites in the early part of the project is also blamed for communication problems and slow project startup. These factors contributed to the fact that it took so long to get COF projects off the ground. The COF comprised a two-phase approach: • ICB - Capacitating ROCs initially by deploying COF grant funding for ROC institutional capacity building (typically support to procure office equipment, transport, staff and community level training, etc), • COF - offering support. for infrastructure projects. This was not ideal, as capacity-building is best done in conjunction with hands on work. Consequently, it did not often reap the expected benefits . There is some debate as to whether the IC employed the "right" kind of professionals for providing assistance to ROCs in the technical aspects of COF project proposals, and whether the IC fully anticipated just how much of this role they would eventually have to take on, given CA's lack of capacity and the declining technical capacity of the NGOs. Consequently, the qualifications of IC project development officers were of a "grant making and administration" instead of a "hands on" practical and vocational nature, and may not have sufficiently matched the requirement for field implementation. To their credit, OA adapted to this situation, employing a seasoned industry insider as a project development officer, after which COF community projects started to roll. In fact, the small grant window facility emanated also in this period (sadly towards the end of NRM II). The project design did not anticipate this wide role for the IC and its staff did not ideally match the requirements of field-based projects. It also under-estimated the need for technical engineering-type capacity, the match between this requirement and CCG capacities, and the declining ability of CCG members to provide the anticipated technical expertise. This had some impact on slow start of community project funding (though not ICB projects). The weak capacity of engineers and technical staff at ROC level compounded the delays. The technical proposals on eco-tourism required sound engineering, architectural design, siting, and marketing input for such projects to comply with the requirements of USAIO grant mechanisms. In its comments OA notes that the first two Project Managers were "skilled in community-based project development, interagency collaboration and training - pretty much the combination of skills needed in the early stages of the COF development and the need to deal with interagency collaboration and CA institutional development issues. About the time that engineering, tourism and strong management skills were most needed, [a person with tourism and engineering experience] was hired, initially under a short term TA arrangement and soon after as a full time employee. that It would be hard to argue that these were not the right kind of people for the COF". Accepting the complexities of the start-up phase, we nevertheless cannot easily find the evidence that Project Managers got the COF working except in its latter stages. Were the challenge of the task, and therefore staffing requirements, under-estimated? OA notes, "it was not originally envisaged that the IC would have to provide hands on practical service". The wildlife utilisation and safari industry in Zimbabwe has some of the most adaptable, innovative and successful business practitioners and technical safari experts in tourism infrastructure in Southern Africa. And yet the IC (perhaps because of CA policies) addressed the issue of establishing direct contact with ZATSO (the leading association in the tourism industry) inadequately, to establish synergies between community infrastructure projects (particularly eco-tourism) and the tourism sector. The IC could have hired expertise directly from the industry to develop ecotourism enterprises. ZATSO and its membership should NRMPFINA LREPORT 29JUL Y2003 109 also have been used as partners in development. This was suggested (e.g. by USAID) but never followed up by CA. DA confirms "short-term TA specialists were admittedly a mixed bag, ranging from solid types ... to other more university types and others with less hands on/relevant experience. It should also be noted that there was considerable pressure/oversight (by CA and GOZ) in the hiring process to spread the goodies around, give preference where reasonable to indigenous folks, etc. The IC was particularly careful of its role in this regard since there was the perception by CA and the GOZ that favouritism might be occurring in the TA hiring business. In fact, The CA Director at one time was actively promoting the notion that the IC was up to no good in this regard. So it is not surprising that finding the highest quality of TA was not always feasible given the above circumstances." DA also distances itself from the choice of consultants and explains: "The system used was for each ROC to source their consultant through competitive means. A database of consultants was kept by DAICA for reference by the ROC's if they so wished. In all cases the ROC would propose their choice of consultant to be approved by OAICA. If the choice of consultants was wrong, the ROC also had a significant part to play in this and should therefore accept part of the responsibility. Consultants were not 'forced on' nor 'allocated' from Harare." d) Bllildillg the Capacity o/CAMPFIRE Associatioll The Institutional Contractor was also responsible for training and capacity building of the CA. Most of the training of the financial section of CA was on-the-job-training, but with high staff turnover it is difficult to see much impact. In general, CA was not willing or able to capitalize on the capacity building efforts of the Institutional Contractor. Although TAwas provided by the IC for personnel development, HRD and staff performance appraisal, the quality of staff currently employed by CA does not reflect much impact of any training or staff upgrading having been provided. The change in CA leadership between 1998-2002 was also a major factor. All in all, IC has been a good gatekeeper, assessing technical and viability requirements and accountability where the USAIO manuals called for this. However, its role as a controlling force and messenger was also frustrating, as few of other PMT members provided similarly rigorous technical scrutiny of projects and it was often the lone voice saying 'no'. In CA, it faced a major challenge in building a sustainable NGO. And, it must be said, it failed, largely due to various externalities and role conflicts. For example, and a design flaw in the project, was the role conflict between CA's capacity building and controlling roles. Its mandate of build CA's capacity was not always encouraged by the latter. Capacity building is a jOint operation for the mentor and the grantee. For reasons of role conflict (i.e. OA was supposed to be a grant gatekeeper and to provide capacity-building) and of personalities, this relationship was not as productive as it could have been through much of the project period. The complicated relationship between CA and DA really compromised the effectiveness of DA operations especially, but not only, with the CA capacity building and sustainability work. D. ROLE OF USAID, HARARE OFFICE The role of USAID, through Strategic Objective Team 1, was the day-to-day monitoring of NRMP implementation and keeping USAID aware of the Program's general implementation status. Specifically, roles included: • Preparation of project documentation, amendments, agreements and implementation letters. • In coordination with PMT, manage and monitor all grants to GoZ agencies. • With BRC and PMT, contract and manage final Program impact evaluation. • Routine USAIO programme correspondence. • Provide material oversight for the Institutional Contractor. • Monitor Program implementation, including adherence to USAID regulations and progress towards stated objectives. USAIO personnel were highly committed to the CAMPFIRE programme, and played an increasingly active role, especially in the PMT and concerning the approval of individual COF projects. USAID was placed in a difficult position following the demise. of the CCG, and the increasing unacceptability of devolution in the Ministries of Tourism and Local Government (MLGPWNH) in the mid to late 1990s. Even retrospectively, it is difficult to judge whether the decision to fund CAMPFIRE in the face of important programme difficulties NRMPFINALREPORT 29JUL Y200J 110 that compromised project design (Le. the increasing hegemony of CA and alienation of CCG partners; the failure by CA to include producer communities as members) especially at the time of the Mid Term Evaluation was correct. However, the marginalisation of the CCG and civic society generally from the programme, left USAID in an important, and possibly unwelcome, leadership and position. Together with DA, USAID was often in the difficult position of having to argue the unpleasant fact that a project was not viable and should not be funded. USAID became very much part of the CAMPFIRE managerial team, largely by default. While this was an important factor in controlling the project and supporting DA following the decline of the CCG, the question of whether a donor should be sucrl an active participant was raised several times: Should USAID have allowed itself to get into this position? In the absence of the leadership that had initiated the CAMPFIRE programme, USAID can be criticised for not being more visionary and forceful in ensuring project sustainability (yet, it is also often criticised for being too "hands on"). While the sustainability issue was often drawn to the attention of CA, and USAID eventually dealt with it by cutting back funding, perhaps USAID should have been stronger and also more innovative in insisting on approaches that would lead to project sustainability. An important case was the failure to develop any form of endowment or sinking fund mechanism. We have already referred to this recommendation in the Mid Term Review. The clash between the four-year establishment frameworks for endowments in the USAID bureaucracy, and the two-year NRMPII closeout period, also played a role. In the event, although NRMP was periodically extended five years beyond the Mid Term Evaluation, there was never more than a two-year planning horizon. Although considered briefly by USAID, USAID never accepted the Mid Term recommendation (at least for the use of USAID funds) for an endowment/sinking fund (Loken, comments). This reflected bad past regional experiences with similar endowments (e.g., Swaziland - 1998-99), time constraints and fears that such a mechanism would be misused by the CA. Instead, USAID tried to help the CA consider something appropriate for meeting this need with other source funds (e.g. Ellsworth report, Murphree proposals, and, later, the SGI consultancies). However, the CA never really appeared interested in follow-up on the early work (as, basically, per usual, they wanted their own endowment, excluding all others, as laid out in the SGI, and the early consultants were not recommending this). Thus, not only was the design, and therefore sustainability, of the CDF constrained by the conflicts and political pressures mentioned, but USAID was not able or willing to consider innovative suggestions. Seven years on, it is easy to be critical of project design, and we do so in the light of the political constraints mentioned throughout this document. Nevertheless, NRMP II was overly top down, overly constrained by bureaucratic procedure, and with more imagination in the design of the project (e.g. match a component of CDF grants to fiscal devolution), it could have supported devolution far more directly than attempting to pursue devolution In a rather cumbersome manner through RDCs. The use of an endowment or sinking fund could also have extended the life of the funding so it better matched the absorption profile of communities. E. PROJECT MODEL AND PROJECT DESIGN Two Government Ministries and several Zimbabwe-based NGOs worked together to initiate CAMPFIRE. These entities, known first as the CAMPFIRE Collaborative Group and subsequently as CAMPFIRE Service Providers (CSPs), provided considerable assistance to the movement during its first decade of development. The CCG also facilitated the establishment of the CAMPFIRE Association as the lead agency to represent the CAMPFIRE constituency and coordinate/orchestrate the work of CSPs to support CAMPFIRE. USAID funds were to support the strengthening of CA's institutional capabilities and to fund most of the technical inputs of the CSPs. For several reasons, most of which we have already touched upon, the delivery model could have worked better: • CA was correctly posited at the lead coordinating agency, but has remained an RDC-linked organization and has not evolved into a true producer association. • The conversion of CCG members into CSPs removed critical long-term relationship-based facilitation from the programme, as well as diversity and civic peer review. NRMI'FINALREI'ORT 2