Report No. 45781-UC Uganda Sustainable Land Management Public Expenditure ~evrew(SLM PER) November 6,2008 Agriculture and Rural Development Unit (AFTAR) Sustainable Development Department Country Department 1, Uganda Africa Region E Document of the World Bank Uganda Sustainable Land Management Public Expenditure Review (SLM PER) ABBREVATIONS AND ACRONYMS AfDB African Development Bank AGRA Alliance for a Green Revolution in Africa BCC Budget Call Circular CAADP Comprehensive African Agricultural Development Program CA 0 Chief Administrative Oficer CBO Community-based Organization CDD Community-driven Development COMESA Common Market for Eastern and Southern Africa CSIF Country Strategic Investment Framework cso Civil Society Organizations DSIP Development Strategy and Investment Plan DTCE District Technical Planning Committee ENR SIP Environment and Natural Resources Sector Investment Plan FA0 Food and Agriculture Organization GDP Gross Domestic Product GEF Global Environment Facility GHG Greenhouse Gas GoU Government of Uganda IFPRI International Food and Policy Research Institute KARI Kenyan Agricultural Research Institute KRA Key Results Area M&E Monitoring and Evaluation MAAIF Ministry of Agriculture, Animal Industries and Fisheries MDG Millennium Development Goals MDPED Ministry of Finance, Planning and Economic Development MDS&tP Development Strategy and Investment Plan of the Ministry of Agriculture,Animal Industries and Fisheries MEMD Ministry of Energy and Mineral Development MFPED Ministry of Finance, Planning, and Economic Development M L r n Ministry of Land, Housing and Urban Development MLWE Ministry of Land, Water and Environment MoLG Ministry of Local Government MTEF Medium T m Expenditure Framework MTTI Ministry of Trade and Tourism Industry NAADS National Agricultural Advisory Services NDP National Development Plan NEMA National Environmental Management Agency NEPAD New Partnership Tor Africa's Development NFA National Forest Authority NGO Non-Governmental Organization NPF National Forest Plan NRM Natural Resource Management PEAP Poverty Eradication Action Plan PER Public Expenditure Review PMA Plan for Modernization of Agriculture SIP Strategic Investment Program SLM Sustainable Land Management SMU Soil Mapping Unit SSA Sub-SaharanAfrica UGX Ugandan Shilling UNCCD United Nations Convention to Combat Desertification UNDP United Nations Development Program UNFCCC United Nations Framework Convention on Climate Change IJWA Uganda Wildlife Authority WDR World Development Report PEAP Poverty Eradication Action Plan PER Public Expenditure Review SLM Sustainable Land Management SMU Soil Mapping Unit UGX Ugandan Shilling UNDP United Nations Development Program Vice President: Obiageli Ezekwesili Country ManagerDirector: Kundhavi KadiresanIJohn McIntire Sector Manager: Karen McComell Brooks Task Team Leader: Johannes Woelcke CONTENTS EXECUTIVE SUMMARY .......................................................................VI 1 INTRODUCTION . .........................................................................................................1 2 LAND DEGRADATION IN UGANDA: EXTENT. CAUSESAND . IMPACTS .............................................................................................5 3.SLM-RELATEDINSTITUTIONS. POLICIES AND STRATEGIES ..................17 3.1 DEFINITION `SUSTAINABLE LANDMANAGEMENTINUGANDA OF ................................17 3.2 THEINSTITUTIONALLANDSCAPE SUSTAINABLELANDMANAGEMENT FOR .................17 3.3 SLM POLICIES STRATEGIES AND ..............................................................................23 3.4 COORDINA MECHANISMS SLM: OPPORTUNITIES C ~ L L E N G E S TION FOR AND ...........28 4. REVIEW OF SLM PCTBLICEXPENDITURE ...................................................34 4.1 IDENTIFICATIONSLM PORTFOLIO OF ..................................................................... 34 4.2 TRENDS THELEVELOFSLM EXPENDITURE IN ......................................................... 37 4.3 SECTORAL FUNCTIONALCOMPOSITIONSLM EXPENDITURE AND OF .........................40 5. SLM-BASED DEVELOPMENT PATHWAYS FOR LAND DEGRADATION HOTSPOTS IN UGANDA .............................................................................................57 6 POLICY CONCLUSIONS AND RECOMMENDATIONS . ...................................74 ANNEX 1:SECTORALCOMPOSITION OFTHE "SLM SECTOR" AND "SLM RELATED SECTOR" EXPENDITURE. TOTAL FOR 2001/02-2005/06 ................78 ANNEX 2: METHODS FOR PREDICTING SOIL EROSION AND NUTRIENT DEPLETION ...................................................................................................................83 REFERENCES ................................................................................................................99 TABLES Table 1: SLM sector and SLM related sector expenditure. 2001/02.2005/06. UGX million ................................................................................................. 38 Table 2: Sectoralcomposition of the actual "SLM sector" expenditure. 2001/02- 2005106 (in UGX million) ............................................................................ 41 Table 3: Sectoral compositionof the actual "SLM sector" and "SLM related sector" expenditure. 2001102-2005106 (in UGX million)........................................42 Table 4: CSIF componentsand main types of activities for SLM PER in Uganda ....44 Table 5: Impact of selected SLM practices on crop yields in Uganda........................48 Table 6: Indicative costs of producing maize by using different SLM technologies in Uganda. in UGX per ha ............................................................................49 Table 7: SLM-based developmentpathways for land degradation hotspots ..............67 Table 8: Involvementof the parliament in the budget process and time line .............88 Table 9: Structureof Mukono District budget. in percent of total ...........................................................................................................94 Table 10: Perceived influenceof different actors on district and sub-countybudget (Mukono District) ......................................................................................... 97 FIGURES Figure 1: Steps for CSIFpreparation.............................................................................. 2 Figure 2: Severity and extent of soil nutrient depletion and soil erosion in Uganda .....7 Figure 3: Land use in Uganda ......................................................................................... 9 Figure 4: Institutional SLM Landscape........................................................................22 Figure 5: Decision tree of the identification of the SLM expenditureportfolio ..........36 Figure 6: Actual "SLM sector" and "SLM related sector" expenditure. 2001102- 2005106. UGX million .................................................................................. 38 Figure 7: Government and donor contribution to "SLM sector" and "SLM related sector" expenditure. actual spending. 2001102~2005106.UGX million........40 Figure 8: Total SLM andpotential SLM expenditure. 2001102 -2005106. UGX million .................................................................................................43 Figure 9: Compositionof the planned SLM expenditure of MAAIF. 2005106~2007108.UGX million....................................................................46 Figure 10: Compositionof the actual SLM expenditureof MAAIF. 2001102~2005106.UGX million....................................................................46 Figure 11: Functional compositionof "SLM sector" and "SLM related sector" actual expenditureunder CSIF. 2001102~2005106.UGX million ...........................47 Figure 12: Cost of replenishingsoil nutrient depleted in Uganda, US$ per ha per year..........................................................................................................51 Figure 13: Targeting of the SLM expenditure to soil nutrient depletion hotspots. Uganda ..........................................................................................................52 Figure 14: Targeting of the SLM expenditure to soil erosion hotspots. Uganda ...........53 Figure 15: Incidence of poverty and its distribution in Uganda.................................. 54 Figure 16: Poverty head count and cost of replenishing depleted nutrients...................56 Figure 17: Land degradationhotspots and developmentdomains in Uganda................58 ACKNOWLEDGEMENTS This Uganda Sustainable Land Management Public Expenditure Review was carried out by a team from the World Bank and the International Food Policy Research Institute (IFPRI). The World Bank task team included Johannes Woelcke (Task Team Leader and Economist, AFTAR), Sergiy Zorya (Economist, AFTAR), Nathalie Johnson (Senior Environmental Specialist, AFTEN), Stephen Danyo (Natural Resources Management Specialist, AFTEN), Meseret Kebede (Program Assistant, AFTAR), and Gladys Akurut Alupo (Team Assistant, AFMUG). The IFPRI team included Ephraim Wlonya (Research Fellow), Regina Birner (Senior Research Fellow), John L. Pender (Senior Research Fellow), and Edward Kato (ResearchAssistant). David Abura (Ministry of Tourism, Trade, and Industry) helped with data collection and data analysis. Stephen Muwaya (MAAIF, TerrAfrica Focal Point), Boaz Blackie Keizire (MAAIF, CAADP Focal Point), Justin Ecaat (UNDP, Environmental Specialist), and Matthias Magunda (NARO, Kawanda Agricultural Research Institute) provided comments and feedback. Kirk E. Hamilton (Lead Environmental Economist, ENV) and Peter A. Dewees (Lead Environmental Specialist, ECSSD) served as Peer Reviewers. Madhur Gautam (Lead Economist, AFTAR) and Wilson Onyang Odwongo (Rural Development Specialist, AFTAR) provided advice and comments throughout the preparation of the report. John McIntire (Country Director, Tanzania and Uganda), Karen McConnell Brooks (Sector Manager, AFTAR), Marjory-Anne Bromhead (Sector Manager, AFTEN), Christine E. Cornelius (Program Coordinator, AFTAR), Stephen Mink (Lead Economist, AFTSN), and Christophe Crepin (Lead Environmental Specialist, AFTEN) supported the study and ensured that resources were available for its implementation. The team would also like to acknowledge the financial support provided by TerrAfrica. EXECUTIVE SUMMARY Introduction and Objectives 1. This report summarizes the findings of the Uganda Sustainable Land Management Public Expenditure Review (SLM PER). The SLM PER was undertaken to achieve six main objectives: (i) establish a robust data base on SLM-related public expenditure that can support credible empirical analysis; (ii) develop a sound methodology for conducting SLM PERs, which could guide similar work in the future; (iii) analyze the level and composition of SLM spending in the recent past; (iv) identify potential entry points for public support based on the concept of development pathways; (v) understand the institutional arrangements for coordination on SLM issues; and (vi) draw policy recommendations for the land use sector. 2. Overall the SLM PER aims at informing the current policy dialogue with and within Uganda and thereby contributing to the development of a common national strategy for the efficient and effective scale-up of SLM interventions. The review is expected to provide important input to: (i) the on-going Comprehensive Afica Agriculture Development Program (CAADP) process, including the SLM Country Strategic Investment Framework (CSIF); and (ii) the Ugandan SLM Country Program that will be supported by the World Bank and the Global Environment Facility (GEF) through TenAfrica. Uganda is preparing a CSIF with the goal of establishing a country- led operational roadrnap for developing a sequenced program of SLM interventions. The PER is among the recommended diagnostic tools for CSlF preparation and SLM awareness building. In general, this analytical work complements the on-going AgriculturePER in Uganda. 3. Low actual disbursement of SLM expenditure and the significant donor contributions result from the lack of clarity on what the priority programs supported by the public sector should be. While donor contributions are certainly useful, particular interests may have dominated the field in the absence of clarity within the Government. This study makes an effort to help the Government of Uganda (GoU) to achieve more coherence in its own understanding of priorities in order to manage the collaboration with the donors and fill in the gaps. Main Findings 4. Five major land degradation hotspots have been identified in Uganda. They include the Southwestern Highlands, Lake Victoria Crescent Region, Northwestern Regions, Eastern Highlands, and the Southwestern Cattle Corridor. It is estimated that nitrogen, potassium and phosphorus balances are less than -85, -75, -10 kglhalyear in these areas. Soil erosion is estimated to be more than 5 tonshectarelyear. 5. The causes of land degradation in Uganda are context-specific and require responses that are integrated into regional economic development concepts. Considered in isolation and countrywide each factor is ambiguous with no clear-cut and definite answer to the land degradation issue. The causes may vary depending on the nature of the specific underlying socio-economic factor as well as the type of land degradation considered. The interplay of various socio-economic factors needs to be acknowledged while designing SLM strategies and allocating public resources for a specific region. The concept of "SLM based development pathways" outlined in chapter 5 can inform the process of identifying entry points for public support based on location- and context-specificcomparativeadvantages. 6. The Government of Uganda has recognized that land degradation constitutes an important constraint to economic growth and environmentalsustainability. SLM is considered to be crucial to reach economic growth and poverty reduction targets. Strategies and investment plans integrate SLM systematically in their objectives and central pillars. Government plans provide an economic rational for public investments in sustainable land management and clearly define the role of the private sector vs. the public sector. 7. In spite of the severity of land degradation in Uganda, SLM expenditure was comparativelylow between 2001 and 2005. Total SLM expenditurewas 0.28 percent of total budget expenditure and 0.13 percent of total GDP. Adding "SLM related sector" expenditure increases the share of the SLM expenditure in total budget to a still low 1.15percent. For comparison, the public spending on the agriculture sector as a share of total budget was 4.3 percent during 2001102-2005106. Most fimds for SLM have been provided by donors. During 2001-2005, the share of donor resources in the "SLM sector" expenditure was 83 percent. 8. The actual disbursement of SLM expenditure was only 42 percent of the planned expenditure. Since this problem is not only experienced with regard to SLM expenditure, the GoU would need to strategically review the common procurement/disbursement procedures and assist programs and projects to address these issues. Furthermore, off-budget expenditure, not recorded in the official budgets, should be monitored and reported on a regular basis to inform the policy makers and other stakeholdersabout the level and compositionof all public expenditures. 9. Nevertheless, the low expenditure was relatively well targeted to land degradation hotspots and was used to finance public goods. Looking at spatial targeting, SLM expenditure was higher in areas with severe soil nutrient depletion and soil erosion. Moreover, SLM expenditure was used to provide public rather than private goods, such as capacity building, enabling environment, and research. The exceptions are the Eastern Highlands, and the area around Lake Albert, where SLM expenditure was low in spite of severe land degradation.It should also be noted that a couple of districts in Northern and Central Uganda, with relatively low degree of soil nutrient depletion and soil erosion, received significantpublic resources to address SLM issues. 10. Spatial targeting needs to acknowledge that land degradation hotspots have limited overlaps with poverty hotspots. Thus, allocating resources to poverty hotspots will not automatically reverse severe land degradation. While the highest incidence of poverty is found in the North and Northeastern Regions, severe land degradation can be observed in the regions mentioned above. Accordingly, only a few districts in vii Northwestern and Southwestern Uganda fall into both categories of land degradation hotspots and poverty hotspots. However, expenditure designed to improve SLM can still be pro-poor. Many land degradation hotspots are highly populated areas and home to many poor people. This implies that adequate SLM investments in areas with severe land degradationmay lead to growth and poverty reduction. 11. In terms of actual expenditure, government spending on the environmental agenda associated with SLM was significantly higher than on the SLM productivity agenda. A direct prevention of land degradation to avoid productivity losses has not been the prominent priority of SLM interventions on the ground; greater focus was placed on environmental issues related to management of watersheds and protected forest areas. Thc composition of expenditure indicates that at the operational level the GoU has likely considered land degradation problems as a part of the broader environmental agenda rather than a constraint to productive land capacity and a prerequisite for long-term agricultural and economic growth in the country. 12. Actual expenditure for scaling-up SLM activities are low, in spite of being identified as a priority area in the investment plans. One possible explanation for this mismatch might be that limited knowledge exist which SLM practices and technologies work best under which circumstances. Another explanation might be that at the operational level the SLM agenda remains disintegrated from the overall agricultural (land) productivity agenda. 13. Expenditure for monitoring and evaluation (M&E) have not resulted in the information needed to further strengthen the case for SLM and to scale-up SLM activities. Little knowledge exists about outcomes and impacts of public SLM spending. So far, the limited availability of such types of data has been a challenge for ensuring an adequate budget allocation for SLM in the budget process. Even if the urgency of SLM problems can be made clear in budget negotiation, making the case for an increased budget depends on the ability to offer convincing solutions and to monitor SLM indicators. 14. Uganda's SLM-relatedbudget managementprocess has many positive features to ensure adequate coordination and transparency. The process is designed to ensure coordination among different ministries and agencies that share mandates for the same sector, which is important for SLM issues. Moreover, the process is aimed to provide transparency from early stages of the budget preparation process onwards, and to involve stakeholders from the private sector and civil society in the process of budget preparation. In hct, considerable resources are spent to organize budget conferences with wide participation at the national, regional and local levels. 15. However, some important challenges regarding the budget preparation process remain, both at the national and local level. They include: (i) the declining discipline of keeping time-lines; (ii) the role of indicative budget ceilings and budget allocation processes; (iii) the limited information of "value-for-money" in the Budget Framework Papers; and (iv) the limited reflection of priorities of civil society and farmer organizations in budget decisions (see Annex 3 and the Agriculture PER for more details). Key Policy Recommendations 16. This section makes some key policy recommendationson how to increase allocative efficiency and effectiveness of public SLM expenditure in the short to medium term and how to identify entry points for public support. In general, the strategy could be to undertake some immediate measures to increase the efficiency of existing resources. In addition, some practical recommendations and guidelines are identified which are relevant for medium-term decisions. Of outstanding importance, however, is rigorous monitoring and impact evaluation. M&E as the most importantstep towards increased efficiencvand effectiveness: 17. Strengthening M&E is crucial to make a stronger case for SLM, inform the allocation of public resources in the future, and is a prerequisite for scaling-up SLM activities. In negotiations regarding strategic priority setting and budget allocation, it is important to show: (i) which interventions work on the ground to solve socio-economic and environmentalproblems; (ii) how much they cost; and (iii) what rates of return they have in terms of contributing not only to environmental sustainability, but also productivity growth. So far, the limited availability of such types of data has been a challenge for ensuring an adequatebudget allocation for SLM in the budget process. 18. Immediate action could be taken to assess the impacts of past and on-going SLM projects based on availablesurvey data. This would be the fastest way to provide the urgently needed evidence of SLM outcomes and impacts. Opportunities should be explored to base the evaluation on various plot, household and cornmunity-level surveys conducted by the Uganda Bureau of Statistics (UBOS), International Food and Policy Research Institute (IFPRI), and others in the recent past. These institutions did a series of surveys since 2000 and they may constitute a good basis for evaluation of SLM impacts. 19. The development of a country-wide SLM M&E System could start immediately. Initial steps may include: (i) a readiness assessment to clarify roles and responsibilities, to analyze the quality of existing structures and to identify capacity building needs; and (ii) finding an agreement on the outcomes to monitor and evaluate and their performance indicators. Designing and sustaining a sound results-based M&E system (including impact evaluation) is a complex task, particularly with regard to SLM. Identification of an appropriate institutional set-up, data collection and analysis require strong analytical skills. Hence, the development of a M&E system for SLM should not be outsourced completely to external experts, but rather conducted in a collaborative and participatorymanner in order to build sustainablein-country capacity. 20. The following key aspects should be considered while designing SLM M&E systems and conducting rigorous impact evaluation: (i) financial attractiveness of SLM technologies and practices for land users is a necessary condition for scaling-up and should be captured as a key element; (ii) explicit consideration of externalities may help strengthen the case for public SLM support and design payment for environmental services systems; (iii) spatial scaling-up of SLM in land degradation hotspots on pilot basis should be accompanied by M&E implementation from the start; (iv) the M&E system should be designed in a way that can help to identify the most efficient and effectivetypes of public SLM interventions. Short-term measures to increase efficiencvand effectivenessofpublic SLM resources: 21. Closing the gap between budgeted and actual expenditure can significantly increase public SLM spending in the short run. As mentioned in the findings above, actual disbursement was only 42 percent of planned expenditure. Reaching the full ceiling of allocated SLM funds would have already increased actual expenditure by Ugandan Shilling (UGX) 19billion per year between 2001 and 2005. To address this gap - an issue which is common across many sectors in Uganda - a strategic review of procurement and disbursement processes and procedures would be required. Clarification on the priorities for public support may further increase disbursement (see section 4.2 for details). 22. Spatial targeting of SLM expenditure can be further improved by focusing on land degradationhotspots as the main allocation criteria. While public resources have been targeted relatively well spatially, there is still room to increase their effectiveness and efficiency. The Eastern Highlands, the area around Lake Albert, and small pockets in the Central Region have received low levels of public SLM resources in spite of severe land degradation. A couple of districts in Northern and Central Uganda with relatively low degree of soil nutrient depletion and soil erosion received significantpublic resources to address SLM issues. If allocations would be revised accordingly, the chances of successfullypreventing and reducing land degradationwould increase(see section4.3). 23. Various measures can be implemented in the short-term to reduce the mismatch between the goals as expressed in sectoral investment plans and actual expenditure. Actual expenditure reveals that the concept of placing SLM in the productivity and economic growth agenda as envisaged by relevant strategies and investment plans has not yet been translated into operational reality. The following activities could addressthis issue: Developing a shared understanding of SLM at the policy and operational level. This SLM concept should be as operationally relevant as possible and take into account country-specific conditions. Further, the concept should clearly acknowledge the role of SLM with regard to productivity and economic growth. The concept outlined in chapter 4 may inform this process. Integrating SLM issues in budget planning documents. These would include the upcoming National Development Plan, Budget Framework Papers, MAAIF's Development Strategy and Investment Plan (DSIP), and all relevant Sectoral Investment Plans. NEMA has developed Guidelines for "Mainstreaming Environmental Issues into Budget Framework Papers" (NEMA, 2007). This mainstreaming provision is potentially a very powerful tool to address cross- cutting SLM issues in the budget process. However, since the mechanism is new, it is importantto assess its effectiveness(see chapter 3.4). Strengthening the role of communities and SLM-related Civil Society Organizations (CSOs) in the budget process, as well as in project preparation and implementation. Farmers and communities have repeatedly raised their concerns with regard to land degradation. As the experience world-wide has shown, civil societyparticipation can play a key role in bringing sustainabilityissues on to the political agenda. Yet, civil society organizations in Uganda are concerned about the problem of "token participation" with regard to their involvement. To address this problem, it may be useful to clarify the role of civil society participation in different processes, and to improve accountability mechanisms for participation at the same time (see chapter 3.4). Medium-term measures to increase efficiency and effectiveness of public SLM resources: 24. Systematic integration into major land productivity programs may constitute a low cost option to scale-up SLM spatially, in particular in the land degradation hotspots. Appropriate up-scaling mechanisms and vehicles have not been identified or utilized yet. Such vehicles may include all major productivity programs, such as agricultural advisory services (National Agricultural Advisory Services, NAADS). agricultural research (National Agricultural Research Organization, NARO), agricultural land management, and large-scale watershed approaches. Regarding the NAADS program, for example, the integration of SLM in agricultural advisory services should become an effective and efficient way of helping farmers to get acquainted with the results of SLM research (generated by NARO and others), understand the reasons and outcomes of land degradation, and adopt and maintain profitable SLM practices. Currently most agricultural projects that seek to disseminate farm technologies have not explicitly included SLM elements, resulting in limited success to prevent and mitigate land degradation and thus bring a long-term sustainability to the current farm techniques. However, such programs would constitute an appropriate vehicle for scaling-up sound management of natural resources (see section 3.4.3 and chapter 4). 25. Public expenditure should be used mainly to encourage more private SLM investments. The private sector is expected to make investments in profit-generating private goods, such as in inputs needed to adopt and maintain financially attractive SI,M practices. The public sector's role is to create an environment that favors these private investments. Examples for public sector roles may include: (i) reducing high up-front costs for SI,M practices and technologies by generating and disseminating low-cost solutions via research and advisory services (NAADS and NARO); (ii) promoting community-driven development approaches based on matching grants (e.g. for development of community nurseries for agro-forestry systems); (iii) providing the regulatory and institutional framework for payment for environmental services systems; and (iv) improving access to in-kind and in-cash credit. On-going and upcoming investment operations supported by the World Bank and other development partners, such as N M S and the Environment Sector-wide Approach, constitute opportunities to consider some of the entry points identified in this study and mainstream SLM in the land productivity agenda. 26. The identification of entry points for efficient and effective SLM public expenditure can be based on the concept of socio-economic development pathways. Proximate and underlying causes of land degradation in Uganda are complex and context-specific. This implies that addressing these causes and successfully promoting SLM in Uganda through public expenditure, requires a spatially disaggregated and location-specific approach. While SLM public expenditure target l a d dtprad~tih~' hotspots relatively well fiom a spatial perspective, they are not directly integrated in socio-economic development pathways for rural areas. These pathways should be based on comparative advantages that exist in particular locations. A wide range of public investments are needed to realize these advantages, SLM support being an integral part of them. Much public action aimed at improving land management focuses on influencing household or community adoption of particular technologies. Yet this may be ineffective if the technologies and practices are not suited to the development pathways that have potential in a given location. Detailed recommendations on priorities for public investments and policies for each land degradation hotspot are summarized in Table 7 (chapter 5), including public support to create an enabling environment. 27. While land use sectors contribute to land degradation, the answer is not to slow their growth, but to seek more sustainable production systems through public and private investments and to enhance the sectors' provision of environmental services. In Uganda, land degradation is most severe in areas with high ago-ecological potential and often caused by high land use intensity. Given the crucial role of the land use sectors for economic growth and poverty reduction and given the limited scope for bringing additional land into production, the response should not be to slow the productivity growth in these sectors but to base the urgently needed smallholder productivity revolution in Uganda on a technology change that systematically integrates SLM aspects. While the major role of the public sector is to crowd in private investments by providing public goods, financial attractiveness of SLM practices and technologies is a key necessary condition for adoption at farm and community level. Integrated nutrient management is among the most promising technical options, which are financially attractive to the farmers and reverse land degradation. 28. Increased public attention on climate change could be used to further strengthen the focus on SLM and tap into additional financial resources more strategically. It is widely acknowledged that SLM has an important role to play with regard to both climate change adaptation and mitigation. Hence, SLM offers the rather unique opportunities to use synergies between adaptation and mitigation and access additional financing. Currently fhding for adaptation is available under three different Funds of the United Nations Framework Convention on Climate Change (UNFCCC) and GEF. While the resources available are limited currently, it can be expected that funds will increase in the future. In addition, efforts can be made to access carbon financeunder the Clean Development Mechanism of the Kyoto Protocol and the voluntary market. 1. INTRODUCTION 29. Land degradation, low and declining land use sector productivity, and poverty are severe and interrelated problems in Uganda. Although Uganda's soils were once considered to be among the most fertile in the tropics (Chenerey, 1960),problems of soil nutrient depletion, soil erosion, and other manifestations of land degradation appear to be increasing. Land degradation contributes to low and in many cases declining natural resource productivity in Uganda. Particularly severe are the consequences for the agricultural and forestry sectors, since their productive capacity largely depends on the productive capacity of land resources. Farmers' yields are typically less than one-third of potential yields from research stations, and yields of most major crops have been stagnant or declining since the early 1990s (Deininger and Okidi, 2001). 30. Land degradation has multifaceted economic and environmental consequences. Onsite degradation of natural capital has direct impacts on agricultural productivity because it undermines the basis for future agricultural production through soil erosion and nutrient depletion. However, due to externalities, degradation of agricultural land also affects downstream users of natural resources, including other sectors. Off-site effects originating from the agricultural sector include: water pollution, reservoir siltation from soil erosion, mining of groundwater aquifers, deforestation and the loss of biodiversity. Loss of forest cover and degradation of the remaining forest resource base undermine forest's environmental services and value of biodiversity, decrease the availability and value of forest products, undermine subsistenceneeds of the rural population and decrease the attractivenessof forest-based tourism. 31. The justification for public sector investment in SLM is mainly based on market failures, such as externalitiesand non-excludability of users, and the public good argument. With regard to agricultural land management, for example, decisions and practices applied upstream affect the volume and quality of water available to farmers downstream and the extent of siltation of water sources and reservoirs. In addition, as mentioned above, land management has an impact on Greenhouse Gas (GHG) emissions and biodiversity. Almost anyone can obtain and use SLM related-knowledge, especially in the form of new technologies and practices (such as soil and water conservation), so there is little impetus for the private sector to develop or provide it. These market failures can be overcome through regulations and investments in public goods. In addition, as the natural resources used in agriculture and forestry are often public or common pool goods that can be used by any and all people, much of the investmentin sustaining or improving natural resources should be co-financed by the public sector. At the farm level, rural households will invest in improving natural resources if it is profitable, although the technologies and approaches that are used are often based on those developed by the public sector. 32. Against this background, the Sustainable Land Management Public Expenditure Review (SLM PER) was undertaken to achieve six main objectives:(i) establish a robust data base on SLM-relatedpublic expenditurethat can support credible empirical analysis; (ii) develop a sound methodology for conducting SLM PERs which could guide similar work in the future; (iii) analyze the level and composition of SLM spending in the recent past; (iv) identifypotential entry points for public support based on the concept of development pathways; (v) understand the institutional arrangements for coordination on SLM issues; and (vi) draw policy recommendations for the land use sector. Figure 1: Steps for CSIF preparation ( 7 ~ ~ ~ ~ ~ a r O a o l l a n r m u x r l m n d 1 0 p o p k ~ . v-part-+anla(mentpvtnerrdomn * +NGOI+hmaassaahs+pNaC- Source:WorldBank (2006). 33. Overall, the SLM PER aims to inform the current policy dialogue with and within Uganda and thereby contribute to the Government's on-going development of a common national approach for up-scaling SLM. The review is expected to provide important input to (i) the on-going CAADP process, including the SLM Country Strategic Investment Framework (CSIF); and (ii) the Ugandan SLM Country Program that will be supported by the World Bank and GEF through TerrAfrica's Strategic Investment Program (SIP).' Uganda has expressed its SIP is the Strategic Investment Program for SLM in Sub-Saharan Africa. This is a multi-agency investment umbrella led by the World Bank and NEPAD that seeks to scale up SLM approaches in the context of the TenAfrica partnership and in line with NEPAD's regional agriculture and environment programs. Additional donor agencies include ADB, FAO, IFAD, UNDP, and UNEP cooperating to more efficiently mobilize $150 million in GEF grants plus nearly $1 billion in co-financing. intent to shift to a program-based SLM approach under the CAADP framework. CAADP prioritizes SLM scale up under its pillar 1,with related land management goals under pillars 2-4. A CAADP roundtable series is being pursued by the Government with support from the New Partnership for Africa's Development (NEPAD), the Common Market for Eastern and Southern Africa (COMESA), the Government, and donors to elaborate the operational priorities. As part of this process, Uganda is preparing a CSIF with the goal of establishing a country-led operational roadmap for developing a sequenced program of SLM interventions. The newly formed "Interministerial Framework for Cooperation on the Development and Implementation of a Country Program on Sustainable Land Management in Uganda" led by MAAIF considers the CSIF as a key tool for long-term implementation of the country's nascent SLM Country Program. As shown in Figure 1, the PER is among the recommended diagnostic tools for CSIF preparation and SLM awarenessbuilding. 34. Public expenditure data used in preparing the SLM PER are based on the standardized Project Information Profiles from the Ministry of Finance, Planning, and Economic Development (MFPED), which include a brief project description, its objectives, expected outputs/outcomes, relevant activities, implementing agency, location, source of financing, amount of planned funds, and some other information. The data is collected across four Pvlinistries, namely Ministry of Agriculture, Animal Industry, and Fisheries (MAAIF); Ministry of Lands, Water and Environment (MLWE); Ministry of Trade and Tourism Industry (MTTI), and Ministry of Energy and Mineral Development (MEMD). The data also includes autonomous public agencies attached to these Ministries, e.g., National Agricultural Research Organization (NARO) and National Agricultural Advisory Services (NAADS); National Forest Authority (NFA) and National Environmental Management Agency (NEMA); and Uganda Wildlife Authority (UWA). 35. Information on the spatial distribution of the severity of land degradation in Uganda was not readily available when the SLM PER work began. We estimated the severity of land degradation using survey data fiom all major agro-ecological zones collected by the International Food Policy Research Institute (IFPRI) in 2000-2003. The total number of households and plots surveyed was 1,426 and 5,391 respectively (with numerous soil samples per plot).2 Econometric methods were used to extrapolate the form and severity of land degradation to areas not covered by the survey. The analysis of SLM-relatedbudget processes is based on a review of the relevant literature and on interviews with approximately 40 representativesof different organizations at the national level and in Mukono District (which was selected as a case study to analyze particular challengesrelated to the budget process at the local l e ~ e l ) . ~ 36. A major challenge constitutes the fact that no methodology for SLM PERs exists that could have sewed as a good practice example (best to the authors' knowledge). However, the approach applied for this study may provide some guidance for futurework. Other challenges are related to the cross-sectoral nature of SLM, which results in increased complexity for data For details of the survey methods, see Nkonya et al. (2004), Nkonya et al. (2005a), and Pender et al. (2004). See also Amex 2 of this report with the summary of survey methods and the overall methodology. 'Mukono District was mainly selected because it is located in one of the identified land degradation hotspots (Lake Victoria Crescent). collection and analysis. The SLM portfolio was identified based on a broadly accepted definition and clear criteria. Recurrent expenditure were not included in the analysis, since a ministry or unit implementingSLM-related activities generally also undertook many other non-SLM related activities. Sinceno specific budget codingis introduced for SLM in Uganda,joint-cost allocation problems make it impossible to estimate recurrent costs (staff wages and operating expenses) attributableto SLM activities. 37. The SLM PER is structured as follows. Chapter 2 begins with the discussion of the extent, severity, and spatial distribution of land degradation. This discussion is followed by an explanation of why addressing land degradation is important in Uganda's development context. From there we turn to the analysis of proximate and underlying causes of land degradation. Chapter 3 begins with defining the SLM concept, followed by a discussion of the SLM-related institutions, policies, and strategies in Uganda. In addition, coordination mechanisms for SLM and their opportunities and challenges will be discussed. In chapter 4, the actual SLM expenditureduring 2001-2005 is analyzed. The chapter begins with the identificationof the SLM portfolio, and then turns to the analysis of magnitude and compositionof SLM expenditure.The allocative efficiency of SLM expenditure is evaluated based on the sectoral composition of spending, alignment with government priorities, and spatial targeting of land degradation hotspots. In chapter 5, a framework is developed which aims at identifjrlng location-specific priorities for SLM-related public intervention in Uganda. This framework is based on the concept of "development pathways" and covers all major land degradation hotspots in Uganda. The spatial analysis of region-specific comparative advantages results in the identification of SLM-based development pathways and potential public interventions. Chapter 6 concludes with the major policy recommendations. 2. LAND DEGRADATION IN UGANDA: EXTENT, CAUSES AND IMPAB."s"'h 2.1 Extent, Severity and Spatial Distribution of Land Degradation 38. Soil nutrient depletion and soil erosion are the major forms of land degradwfiort Eat Uganda and have reached alarming levels. Other forms include soil 0711:? XI.^.':! - , > .. logging, and surface crusting (Zake et al., 1997; Magunda and Majaliwa, 200O) F . rr - 7 , t i i , I severity of land degradation in Uganda is widespread. Slade and Weitz (1 091j - Pstir * . I erosion and nutrient depletion contributed 85 percent of the cost of land $egacildt,r.a 'i: t , s, In some regions of Uganda, 60 to 90 percent of the total land area is reported ro be aittcteti by soil erosion (MNR, 1992). Already in the 1980s, soil fertility mining in Uganda was at !he highest rates among Sub-Sahara African (SSA) countries, with an estimated average anv~a:rate of nutrient depletion of 70 kg nitrogen (N), phosphorus (P), and potassiunl (K r pt-..1.1- i?:.: (Smaling et al., 1993). Even higher rates of soil nutrient depletion wei..: ohsewed I:? and Kaizzi (1998) for several farming systems in central and eastern TJg?r?dsin : ,' 1131 i 39. What is the spatial distribution of the severity of land degradatlow in Cpanmtt;~: i'rrs- information is not readily available. Available data or analyses either measured land &b~sda:ion in only some parts of the country (e.g., Wortmann and Kaizzi, 1998) or did not gi1.e the spatial. distribution of land degradation (Smaling et al., 1993). Therefore, land degactatin-iow: I ; estimated using survey data from all major agro-ecological zones coI1ectt:d 1); ti r ~ s , . , a . Food Policy Research Institute (IFPRI) in 2000-2003. The total nun~b,i;rr1 ~ . II ~ ~ S L i. I ~ . surveyed was 1,426 and 5,391 respectively (with numerous soil samples Fes n;b!). ' . methods were used to extrapolate the form and severity of land degradation to areas ilor , by the survey (for details on the methodology see Annex 2). Since soil nutrient depletiar aria se.11 erosion are the major forms of Iand degradation in Uganda, they are used as indicators fia 'anand degradation. - 40. The hotspots of land degradation are In the Southwesrsrm ITighiaa~dl-,a 1 . G r--l Crescent Region, Northwestern Regions, Eastern Highlands, and the ?-ogaakcv --: f%cs 2 Corridor. In all regions nitrogen, potassium and phosphorus balances are less dlan -$:, - :. -I, kg/ha/year. This implies that in these regions the total nutrient extraction is more then 170 kg/ha/year (the estimated spatial distribution and the severity of land degradation are shown in Figure 2). This value is more than 280 times the average application rate of 0.6 kg per !m;x of nutrients through chemical fertilizer in TJganda (Morris et al. 2007). According to r.:rri ;i + ; >rjy: by Menao and Baanante (2006) the average soil nutnent extraction is ,,I ;girii :,, erosion - the other maj~;forrn of land degradation in IJganda -- :s greatel. ;. . these five land degradation hotspots. 41. The cattle corridor, which runs from Karamoja area in northeastern Uganda to the southwestern districts of Mbarara and Ntungamo, is widely perceived as being of particular concern with regard to land degradation. Figure 2 illustrates that in particulzr thr southwestern part of the corridor experiences high degree of soil nutrient depleiiar! *:G'"-.~~ - " reaches moderate to high levels in this part. The northeastern pa.;< c::f the csr-i:! r. For deta~lsof the survey methods, see Nkonya et al. (1997), (1999), (2004), Nkonya ~6 a1 (L?OO*:a,t,ant j . i ~ ~ * ea (11. (2004). more moderate degrees of land degradation. The area around Lake Kyoga and smaller patches in easternUganda around Busia have moderate to low soil nutrient depletion and soil erosion. 42. Nutrient depletion and erosion correlate strongly. There are only a few patches with high nutrient depletion and moderate or low erosion and vice versa. This can be expected since soil erosion is one major form of nutrient outflows (in addition to crop harvest, leaching, and gaseous losses). In general, the predicted severity and extent of nutrient depletion and soil erosion are comparable with past estimates (IFDC, 2001; Nkonya et a!., 2005b). Figure 2: Severity and extent of soil nutirient depletion and soil erosion in Uganda Nitrogen Balance 4, Potassium Balance v ; o ' y i 0 L_J-75-0 '-75 No Data r_1 NOData Phosphorus Balance A Soil Loss A >'o ' /jggjhigh(>5) -10 - 0 moderate14 - 5) nNo c -10 Data low (<4) L_1NOData Source:Own estimates. 2.2 TheImportance of Land Degradation in Uganda'sDevelopment Context 43. What are the consequences of land degradation for Uganda's socio-economic development? And why should the Government of Uganda (GoU) allocate public resources to address land degradation problems? This section tries to provide answersto these questions. 44. Land constitutes an essential form of natural capital for various important economic sectors in Uganda. The degradation of this capital most directly affects the agricultural and forestry sectors. The productive capacity of these two sectors is directly linked to the productive capacity of land. Hence, the future of agriculture and forestry is intrinsically tied to better stewardship of the natural resource base, in particular land, on which it depends. However, other sectors such as fisheries, tourism, and energy experience at least indirect effects of a deteriorating land resourcebase. 45. Agriculture and forestry contributeto development in many ways: As an economic activity, agriculture and forestry can be a source of growth for the national economy, a provider of investment opportunities for the private sector, and a prime driver of related industries and the rural non-farm economy (WDR, 2007). The agricultural sector contributes 36 percent to Uganda's GDP and 90 percent of total exports. The composition of the agricultural GDP is dominated by food crops contributing 54percent, followed by cash crops and livestock which contribute 16percent each (MAAIF, 2008). The forestry sector contributes about 6 percent to Uganda's GDP (MWLE, 2001). Forestry provides income through employment or sale of forests products. Fuelwood energy is the major source of energy for domestic cooking, heating and lighting the country. Over 90 percent of Ugandans use fuelwood as their main or only source of energy, consuming 16 million tons each year as domestic firewood and 4 million tons as charcoal (MWLE, 2002). It is estimated that 800,000 m3of timber are used annually for construction, furniture-making, and other manufacture. The value of non-timber products derived fiom forests, such as medicine, craft materials and food, are estimated at about UGX 66 billion annually. In addition, much of the tourism in Uganda is based on forests and their constituent wildlife. The Uganda Wildlife Authority (UWA) revenues from tourism are approximatelyUGX 2.7 billion annually. As a livelihood, 3.8 million households (75 percent of all households in Uganda) are directly engaged in agriculture and 86 percent of the population lives in rural areas. The sector provides jobs for 70 percent of Uganda's labor force, "farm-financed social welfare" when there are urban shocks, and a foundation for viable rural communities. Hence, agricultural performance has a direct impact on poverty incidence. These facts also illustrate the importance of agriculture for food security. Agriculture has done very well over the past 15 years, averaging an annual growth rate of 3.8 percent. In recent years, however, the growth rate has slowed down considerablyraising concerns about the sustainability of the trends in poverty reduction and welfare improvement for the rural population. Within agriculture, the crop sector is dominant with 61 percent of the rural population living in crop farming households. This is important to note since, among all sectors, poverty is the highest among those dependent on crop agriculture. Despite the good past performance, growth in the sector remains volatile, being highly dependant ( - ; I weather using rudimentarytechnology. The forestry sector creates 850,000jobs, about 100,000in the formal sector, the rnaj!~;i:*,~ in fuelwood and charcoal production. In the informal sector, the majority of activity :s :I. ' household fuelwood production. In addition, a large proportion of the rural popl.1X:!L" r depends on forest resources for basic subsistence needs, e.g. fiom farm forestry, or frc~? natural forest and woodlands. Forests provide safety nets against shortages of food. !'uI:' and incomes. As a provider of environmental services, forestry provides a range of envir~im,rnt:i services and biodiversity values, such as greenhouse gas (GHG) mitigation, watersh~,~~ regulation, climate regulation, soil and water conservation, and nutrient cycling. 717e agricultural sector has also the ability to reduce GHG - in particular through SI,M - '0 manage watersheds, and to preserve biodiversity. In both sectors these values LI:-C generally unrecognized and unremunerated. Figure 3: Land use in Uganda Tropical High Other Forest /- 3% Woodlands/ 'il Grassland 25% Source: The Government of the Republic of Uganda (2001). 46. All of the above services depend directly on a sound management of land resources The total land cover ci Uganda is 204,000 krn2. 41 percent of the total land cover is u~lidi.? agricultural production (almost exclusively subsistence farnlland), 25 percent under gjassl,~.~lc:. and 19percent under woodlands (Figure 3). Forests and woodlands cover approximateiy I5 million hectares in. The majority of this area is woodland (81 percent), followed by tropica! !liri forest (19 percent) and forest plantations (1 percent). In addition, there are substantial on-f:i:- - forest resources. The area under subsistence farmland holds an estimated 24 percent of nationill . biomass in the form of scattered trees, forest patches and ago-forestry crops. About 1.9 rnillior ha is currently included in the Permanent Forest Estate. This includes all forest reserve land ar,2 all forested areas in National Parks and Wildlife Reserves. These areas are set aside pem~anenti. for the conservation of biodiversity, protection of environmental services, and the sustainable production of domestic and commercial forest produce. 4 ' Cropland and pastureland make up a very significant share of the total wealth in Uganda. Natural capital accounts for 105percent of the total wealth in Uganda. Produced c:spital, defined as the sum of machinery, equipment, and structures (including urban land), account for only 12percent, and intangible capital for -17 percent. Further breakdown of the ~lduralcapital indicates very high values for land resources. Cropland and pastureland account for 91 percent and forestry (including timber, non-wood forest products and protected areas) account for the remaining 9 percent. Natural capital also accounts for a significant share in neighboring countries, but are not as high as in Uganda. The shares in Kenya, Ethiopia and Rwanda amount to 21 percent, 41 percent, and 36 percent respectively (World Bank, 2006). These figures suggest that managing land resources must be a key part of the development strategy in Uganda. They are a strong argument for the role of land resources in reducing poverty and fighting hunger. Sound managerncnt of these natural resources can support and sustain the wc!fare of countries, such as Uganda, as they move up the development ladder. 48. Quantitative studies confirm high economic costs of land degradation iu Uganda. Zstirnates of the economic costs of soil erosion and nutrient depletion in SSA countries vary between 1percent and 11percent of agricultural GDP annually (Dregne, 1990; Bojo, 1996; Yuben et al., 2007). Dreschel et al. (2001) estimated that the costs of land degradation amount to 6 .I 1percent of agricultural GDP in Uganda annually. In Ethiopia, it is estimated that annual productivity loss from soil degradation is 2-3 percent of agricultural GDP per year (Koehlin, 2U06). A study in Kenya used data fiom long-term experiments conducted at two research stations of the Kenyan Agricultural Research Institute (KARI) to estimate the impact of soil loss an maize yields. Both stations are located in the high potential areas that are comparable to the biophysical environment of some land degradation hotspots in Uganda. The results.indicate that scil loss would lead to a decline of crop yields from 5 tons to less than 1 ton over a period of 50 rears (Nkonya et ul., 2006). 2.3 Proximate and UnderlyingCausesof Land Degradation 49. Previous sections discussed the extent and severity of land degradation, its impacts on i.;ganda7seconomic development and the justification for allocating public resources to address the problem. To design policies and strategies and direct public investments that address the land degradation problem in a sustainable manner, the proximate (direct) causes and underlying (indirect) causes need to be identified arid understood. 23.1 Proximate Causesof Land Degradation in Uganda 50. The proximate causes of land degradation include a range of interrelated biophysical f~ctorsand unsustainable land management practices. Important biophysical factors that affect lznd degradation include topography, land cover, climate, and soil erodibility. The magnitude of soil erosion is a hction of slope length and steepness. Sloped lands are p~sticularlyvulnerable to soil erosion if they have inadequate vegetative cover and no physical barriers to runoff. The densely populated areas on steep slopes of the southwestern and eastern highlands (including parts of Kabale, Kisoro, Bundibugyo, Kasese, Kabarole, Kapchowa, and ~Mbaledistricts) are severely affected by soil erosion (Magunda and Tenywa, 1999). However, hi;rhlands in most cases experience good rains, which contribute to good vegetative cover and hiy.h soil organic matter. These attributes improve the water and soil holding capacity, reduce SII; face n~noffand increase the soil physical stability, all of which help to reduce soil erosion (V:)r>rtrl.~anal., 2000). c:t 5 1 Climate is also an important biophysical factor that affects the extent and severity of land degradation in Uganda. Voortman et al. (2000) note that high temperatures and intense rainstorms in the tropics subject soils to climate-induced degradation. According to Magunda an81 'I'enywa (1999) rainfall intensity is one of the most important determinants of soil erosion in I Tg anda since it is very intense in some parts of the country. Even in drier areas like the cattle co~ridor,rainfall often occurs in intense bursts, and since vegetative cover is poor in these areas, thct soil is exposed to severe water and wind erosion. High and intensive rainfall may cause co~lsiderableleaching, which also leads to land degradation. In high rainfall areas such as the so~~thwesten~highlands, the eastern highlands and the Lake Victoria Crescent region, leaching is a sigrl~facantproblem, especially in sandy and loamy soils (NEMA, 1998). Ssali (2002) also s11t~wcdthat in the central and eastern regions of Uganda, soil acidification and depletion of Ixlsc~caused by leaching are serious land degradation problems. 5% The impacts of the biophysical factors on land degradation are often aggravated by unsa~atainableland management practices. Only a small proportion of smallholder farmers use incargal4c or organic fertilizer, or other fertility management practices and technologies. The ,ur.$xrntof' fertilizer used in Uganda is among the lowest in the world. While Kenya used 32 t.g :ltld Ethiopia 16kg/ha annually between 1996 and 2002, only 0.6 kg/ha were applied in the sane period in Uganda (Jayne et al., 2003). The average for East Africa were 8 kg/ha. 95 percent of ;hetotal fertilizer use in Uganda is by large scale-fanners and tea and sugar estates (Magunda ail(I 'T'en~va,1999). 5 3 'I'raditionally, soils in Uganda were cultivated until crop yields deteriorated to Ilrn:tcceptatPlelevels and the "tired" pieces of land fallowed to restore soil fertility. However, f;~l~?vlngis becoming less common as population pressure increases. Due to extreme land scar-cityin the densely populated areas of the country, such as the southwest highlands, fallowing for one year or more is no longer practiced. Only 6 percent of households used fallow strips in the late 1990s, and the average fallow times decreased from 2.2 years in the late 1980s to 0.7 ~ C TSE in the late 1990s (Pender et al., 2004). 54 B*;arioassoil and water conservation measures were widely practiced prior to the 19"7Os,promoted hy educational programs and often enforced by local adn~inistr.ators.A C < ~iI cbir~ationof several factors (including two decades of political turmoil) led to the neglect or desll-uction of old investments (such as terraces) and discouraged investments in soil cor senation. In addition, cultivation of steep slopes without adequate soil protection has cortribuled to soil erosion, particularly in the densely populated southwestern mountainous rrgtor;. 5" 'osrversianof forest land to farming accounts for a large share of deforestation. As an euamplc, 50 percent of the forest area in and around Lake Mburo National Park was converted to fanning f'rom 1955 to 2000 (Mugisha, 2002). Since over 90 percent of Ugandans depend on he1 wotrd for cooking energy (NEPAD and FAO, 2004), charcoal making and file1 wood are also rnaj11. ~ontributorsof deforestation. The remaining forest resources continue to degrade. In 5:;. @')t&euieair; a . . J :iiaf contribute to land degradation are f i ~ . ~ ~ ~ , ~ ~ifiq~sa,4~;:,:q,,'i- ~ :.:r,ir- ~ e.~;~ ,, ! , - i.: cindi;nc.r of overgrazing in the cattle : .. , .:;:,.I-I. .! ) . . ..''g?r.!ta. i3un tn the red~~ctionland cover, in \ I , , . , . r i::,..- -,.l.. . I . . . ! ,~T,;~~gt,:;~~:l~5: . *'!.!~:ip?<:!:rir~. Fxmcrs a!so burn bushes to encourage ~ C ~ , , , . ~ , , I . ', .. . . , .:" ,,r:?W!~, ,,E'tll';l; tt:.i-l:. . ;, . ; :.IF::- i:iytd for cultivation. Bush burning destroys 5 ; , : .!8: ,. .,,,;;4,...'I, . :' , ~~pv~71.i~pi.:,~;,: :,tw!..<,:8 . , . . :sj~~ii~ist:)il to water and wind erosion n g f ?;I1- .\;:, 1 !~>{.).? ' , . ' ., . . 5'7. 'li'h~.inndf.r.i:/f~~~~: tie!!rad~tifin inelude a range of complex and , . , ? ! , I S . F i : : ~ h c i t , ; ioterreP;?tea$socicr .ei.:ar:t.*,i.li, . . u f i li:,:li*:.. r ptaitgd f??~f@yg;~, I inderstanding their ilnpact is the key . . c I <... r.l.;dress the !and de'gradation problem , . ' , . .',., . .~>".; ; i . , . G . T - . 8 - .$: rCcjcie:;!jl! c:rr{ ~.t;,., i.,,:f.ii . .. .,! : ., t : j ~ i t - . ps:;::ilatitio.gressiae; access to markets, rural fi3P~jrC3 :,: . ' ,:!.~?.ir!!; 1:le::erttralizatirn; privatization of basic I.lrlt:! .: . ,. : .!, . - .. i t : .. t.. . . ,,.. < , . S~:J.VICI: iicji kc; l, I!.::- ::: c ! . ;.,;gr.2.-.!: g t l ~ z ~policy reforms. In the following the a l . - it. .it ., ..!;L .,., .. ;L,;:; ,. l a - . . ::: .. .' :;.:.lii;.:,;~cl...? :Ii~or~:ticalconcepts and available empirical : ;'..... .s&:,... #: 1: t\.jdr:c!rc:: 8 , I:Tg:tlbGrr':. - f.i~i.fctSCS~ from 5 miiPi0n people in 1948 to ; . I A : L " c;; i s : . , - . . 4 . .' ;. :~!!~i>bii ; ~ s ~ 3 ~ ~ 24,2 milji~ppp4.,p!~ :<;h!;' . .i', :;i, : :"!. :'il,: cr::?..ai: popu!atisn growth rate between 2991 and . j g j ' ? \';a; -,v;;,fi c: : I.F :,.,. : .. 1.. ;.. : .. ;. . , I ;!': Lr .- :l-;:$i;L !. . : . . ' - . .t . Z~JO? (!bid). Population pressure is , ! c . ~ t ~ 2 3r x ' . ~ ~ ~ ii .!,,i~ iil 1 . ' i;lainila., tl~eeastern and western highlands . 6 7 ' 1 $1 ,{; f,-: 1>... :.-5 ,, . , , ;,.- ! ! I . , . f31;. . I;, fg>!: : : ::'Oy* 5...- $ t ! , ;; ', :;, I ~ C T ~gr.~)wthIF commonly hypothesized to I ~ : ; ; . ~ ! I ; . ~ , ~I~~:!~!.;:~:;;?..,. I . : l s :hc pressure on arable land, resulting ~ . ; , i L:v:!; : i PiC!*r-:iisi;L . . larid frap?lc~)t;lii.~ir, :i:c:f::.:~.;. ';liS~i~~:~c)cl!: j l : . i ai:d in many cases continuous cultivation of land. . -oil raiilrieat deplel:ion, as more production is sold to .,7 8,3i-c,t1dlilglI~f'~l3~trir::;;j[ 1 - t ; t f ~ ,:';:,$ :i.:i*~'...,.: . . 4 : I ! , ( . ihe nutrients being exported through !?.. , , ' :..,.L,.'.. , c , : : : t : ~ ! ~ i . r ?,-I,? . I , ~ F ? # , ) , ' :' . , , , ,, . , . :'-:)a: t \ ~ l f i t l : ~ rpig phe condition of r X . : .. ,. ~ ~ . ! t : ; k . A :. .%.:, ,. . -..,,JI ! : iezd ro land degradation if it *, , , , .. . : , ;,;$ i . i i i . : r IS ~cl~,cJi;~p~~ljerj 'i,? ;2iF:;li t i :: ...:.:.: a - > i : :ir:w!)er;$!:;+~:T?:':S inay intensify and even improve land ; marlagt.nicr!l ir.rt:snc-fijw i; j. ,;I,! r;r ~ . I . C ~ L<';I !:is ~j! ; !ginally wgued by Boserup (1965; see also , - Pendex t;r (11.. 20i!l, ' 1 l $ i : i : . .:ri[)ij. 127\i~..71~iL;!~~! evidei~ceof impacts of population pressure on .. . ).eqOl,ira-!*,,;.b:i:b'l:i.:r; ' ...'.'. .., (2001) found that , ) I , . * .:,~*+,;.,is P ! . ~ ~ ~ ; I G Y s . Place et / ..1 t*--: CL CFII $ '3- ; .,, , : 1 ,. ; ! x j : , ; ; ;:.. ,'-.!, , . . .., , ,, ' -,:. *.... .>..+ :.!; $3;.1v Size olr t ~ e epltmting, but a positive < I . ~ - ~ I ~ : , ; ~,,,!. . ~ ~ ::,!. , , . . : , . . rrtl! ov!vn i;:nti. Fender el rrl. 42001) found jtl::~~t:,-,-;!it: ' Poverty 60. The proportion of people Ei:.i:?y :;-7:;14 tiif: &,:;1,. : :: i , . . - .. . ..;I. .-i .* ., ' .. tai.<> 1.:. : pi.^"" fi',i+m.ki 56 percent in 1992193 to 31 yerceat in 20ClrE!iI,:., ! ::.: :.I:: 6l.i: .,:!.:t.!fi ;:-:, r c ~ ~ .;:)\..CI;;; : : . , ,_. remains an urgent issue, particularly ~-?llal 6 . . dn.;;j. i j!,:i.- !Ig,, :., i j : j r ' j ; , : .!:':.; r ; c ; i j i . 1 , i gtt: cffci.i ofpoverty on land manageinent is difficult il: i..?::ii!.i. ;,... , .., :,:. "'B:, . -c ill(. I . . z. i -:<:~ J ~ . Y I I ~ C ! ' : , ~ I ,L2 >,., " ability to pay for investments in lard im;..!,t:.:~::,~.,;.~:~I,;.:, ,l,,ii .:, : ~ , . pit ,. i~.+s~,,,:~+~:~; :;j. fmers, which may limit their abj1ir.y ;A, I:!, A!,: ,-..\:. . . .:. .;: i t j ,. f ,;: :.. ?n',l~el {('?:;ti F;. ;:.! : ;is conservation (Pender, 1996; Holdeq ei r:l., ':.'lt41: i i : : . . . . ., , . .- , , .: !. . I , ~iia;,, , ( :.$ ' b :j::t;~]t ~. have more incentive to conserve ';heir iirr;." ',:,: I CiAt,, dl:i:.; * ' . L ' . , *.i ; ; I ,, ' , b , , , , i ; - ,:, ,,.: < ! , , ,;.r,-, ~ and because the opportunity .costs of r.hei; !;; %: : - . ' <,;!( i: ;,)!-. ,,::!., . . . : : ; ; t ~ l : ; d . %b G.,;itL.%. , I , improvement lnay br: lower than the l;lbc;r .:~:jak :,.f. ii:2$.i. -..i..:;;. '::ti :I' :i4::z . ; ; ~ : >:a.i 1~)(I'J; $+, : d i ; ;; Clay et al.,2002). 61. In Ethiopia, Holden and Shiferat'c j2fdlB2: :: t , i : . a t ; a . ,:..:I::. bk;;i.r;?te:hi;.igna h:ad less 3 . i ~ ~ : *.a::! ti- ability to invest in soil and water cssrseh-'t?aticnIB1Z:!N5kr t:.!. :.,'r; :::..r*;, .i:.:: . $ : l!:,sf!;; i:cji?::l!:j~i~es tt! land degradation. Woelcke (2006) folu~d L. ~!:;,i!:i::i.:: l i : ~. ' ]., .,.;, , , t , - , $ * > , 3 i i ' . (.:...- .. * > & , ,,:!' 's:i:alth :f:;;:..;< 1,)t income on adoption of improved !and ma~i,::.rt.n:+;!?t ;? . :.,.;t.;::.:t\ I?I: lad C . .. t , ,:~- %.,.I . . ..>: management and land degradation tll~s;;pp,sr (8I-.. / t : i :.i'::: li, !:, ;i...!I;~:,.:.:blj~<.,: fi~., :T: zlld mla:' vary depending upon the nature of pevi.t1:., ;!:,;:i:..;c .. . , . !L+,:.? , .- * :, !I.:.< % ., ' . . . . I I degradation considered. Access to markets 62. Lack of access to good inbi.ss%ranrtalr.e i * i ? ,:.. v!L.L+. .: d :as t?ae ~LI(+s% significant constraint to subsistence farmers' ncct.aSr??-s ert;rrk.~j,~<;a*: ,'; r 1 i'!s:i :1Ucfr)j ' ;?! i;!Vl..{ 1 Road infrastructure in Uga~,da,dfhouqh ia~rr,ri>i:.~~k:, i:., s ~ i ~: !x ~ , ~ ~ . ; ~ : ~ , : ~ : ~ , ~ i : ; c ~ kjt~ :~< ~ i ~ . i ~~ ~ fil$,r-, 90 percent of Uganda's road nemnric co~:s::;?.i;iii' .:<>,,.t!i,,):I: : . i . i b :,. .: :::P:: .:,:~i: .j-> t)., L.,:QJI,. !.,f' 1.5~ rural feeder roads are impassable durillg rai:;;? ,:rfW. : :i,!, j : .*.#;c :. .I ;;]if; il 4 , .! s, . :J' ! t b m many parts of Uganda hinders the devsiopril~:;!!o! i < > i d t ! ~ . .:;1\5.i.;,!,. .,'L!, .~ ~ ~ ~ S ~ : I i, i~~TI?':( ~ G ~ S F ~ highlands who face most severe land .\degra~,ia.:ic?~: :2u..,;.: 'i::.. . - ;- . !.,wi i;.::,;;. r : , ../ lji't . . these unfavorable conditions, it is estimated ~ b z/ii . ;zjiu:.?i3 j :l:s:i ' . i t : ..: a:.;- I+i'ii ~I::.:tr,~.-!: L i i ; t , i ~ tii. : is transported as head loads, 20 perc:::nr by 1?1:; : . I 1 ,. ...,.:. :- :':,.: :i; : .. ' , -i;~.,t 1; I;:: Ic:i:i~.;Ei,s, ii:, r l 7, i 7 : i t x + a ;I input use, increased participatmn in F:., c i l ~ i -. ..*i.:*r. *,+ , " ,,,* :;4tti81A $,. + A imprc)vements. High costs ~ O Ifert~ii,/er E -: . I * ' , .:,\ .!I 1 ' , L L studies reported low fnancia! ~~.tupi:> 2: ,:1 it, , r :' (. - I , \ ' 1 :, 2004; Woelcke, 2006). 64. With better market access, farmers may be able to shift to the production of highel-, value perishable crops or livestock products, which can also increase returns to input use and offer new land management opportunities. However, better access to markets and transport infrastructure may also lead to more severe degradation of land as farmers may not replenish sufficiently the nutrients they deplete in response to higher market prices. Better access to markets and roads could also create incentives for farmers to cultivate in fragile lands, which in turn could trigger land degradation (LaFrance, 1992). Ruraljinance 65. The absence of a well functioning rural financial system is considered to be a significant obstacle to agricultural development in Uganda. In 2000, 95 percent of households had access to some form of credit, but only about 20 percent of households had access to formal credit. Access varied widely across the country (Nkonya et al., 2004). Lack of credit reduces their ability to acquire and use purchased inputs needed for sustainable agricultural development (Larson and Frisvold, 1996). It also contributes to a short-term perspective of farmers, which fuels overexploitation and degradation of the natural resource base (Pender, 1996; Holden et al., 1998). However, impact of access to credit on land management and the use of inputs are also likely to be ambiguous. Access to credit may not directly translate to purchase of agricultural inputs. Deininger and Okidi (2001) report that only 15percent of loans in 1999 were used to purchase inputs and only 7 percent of loans were used for agricultural investments in land and livestock. The largest share of loans were used to establish non- agricultural enterprises, and for health and education expenditure. Hence, access to credit may not necessarily lead to SLM if other underlying factors of land degradation are not addressed. Land tenure 66. Land tenure security can influence land management since it may affect farmers' incentive or ability to invest in land improvements. Farmers holding land under ir~secuse tenure are less likely to invest in soil and water conservation structures, planting trees and other long-term investments (Feder et al., 1988; Place and Hazell, 1993). In principle, land tenure security can bring direct SLM-rel~izdbenefits through two mechanisms: (i) disincentives to invest in land or capital equipment for working the land can be removed; and (ii) by increasing access to credit, thanks to the use of land as collateral. The latter can remove constraints to, and reduce the costs of both capital for long-term investments and working capital. 67. In Uganda, there are four major land tenure systems: customary, mailo, freehold and leasehold (Republic of Uganda, 1998).The customary land tenure is the most common while the leasehold and freehold tenure systems are the least common (UBOS, 2001). The customary land tenure is a traditional land holding system that is governed by customs, rules and regulations of the community. Holders of land under the customary system do not have a formal land title, btrt generally have secure tenure. Customary land tenure also has some restrictions that are imposed and enforced by clan leaders and elders. These restrictions are likely to affect land management. As an example, landowners may not sell their parcels without consultation with clan leaders and family members. The mailo tenure is predominant in the central region. It was established in 1900 by the British colonial government, when it gave legal land tiles to the royal family and other nobles. 68. It is a common assumption that freehold tenure with full title is the most ~~~~~~e aut: efficient form of tenure and most supportive of SLM. Recent emyiricai evide~lz~ i: challenging this assumption. Nkonya et al. (2004) found insignificant difference. on C~L:UT;Z. plots and freehold plots (with or without a formal title) regarding agriculturd! irria :abL management decisions, productivity and soil erosion. However, a land title was assoc~ated\v 1. greater use of inorganic fertilizer. Another recent study found that some land ~llarirtge~r t.. practices (e.g. use of labor, slash and bum, and incorporation of crop residues) were greare: o . customary plots, and that soil nutrient depletion (especiallyof potassium) was less on c ~ s l m , ~ ~ plots (Nkonya et al., 2005b).~ Local institutionsand extension services 69. Local institutions play a key role in management of land and other natural resources With the beginning of the decentralization process in 1993, local institutions have beel empowered to manage natural resources and the environment in general. District and locc; environmental committees have been formed to enact and enforce environmental and natura: resources ordinances and by-laws (Lind and Cappon, 2001). Decentralization has taken advantage of the role Non-Governmental Organizations (NGOs) and Community-baseti Organizations (CBOs) play in rural areas. These organizations were given representation in the local environmental and natural resource committees. In turn the organizations are required tc: sensitize and assist local people to use sustainable and improved resource management strategie~ and to observe environmental by-laws and other regulations. Empirical studies observed that compliancewith Natural Resource Management (NRM) regulations enactedby local government and other local institutions (such as NGOs and CBOs) was higher than those enacted by highe authorities (Nkonya et al., 2007). 70. With regard to extension services, it was observed that NGOs and programs WELL focus on agriculture and the environment were more likely to provide SLM advisop services than the traditional extension services. The National Agricultural Advisory Service (NAADS) was initiated in 2001 to address the weaknesses of the traditional extension services. NAADS established a decentralizedfarmer-ownedand private sector serviced extension delivery system. Recent studies have showed that NAADS had non-significant impact on SLM (Benin t ? ~ al., 2007). One factor that could be hampering the use of SLM practices and technologier; :s conflicting messages from differentprograms and organizationsinvolved in technical assistance As an example, someNGOs are promoting the use of inorganic fertilizer while others argue I ~ L . they damage the soil (Bashaasha,2001). Policy reforms 71. The major policy reforms that may have affected land managementin Uganda can be summarized in four major categories: SLM-sector policy reforms, international trade and The above fmdings do not mean that tenure insecurity is not a problem in Uganda. According to a shldy ot- ii.1 conflicts in Uganda by Deininger and Castagnini (2004) up to 5 percent of households had a land conflict pendlnz the time of the surveyand a similar share had experienceda conflictduring the prior eight years. fimncial sector reforms, market liberalization and privatization, and decentralization(Nkonya et ell.. 2004):~ ' 2 . International trade and financial sector reforms: The GoU abolished or substantially reduced export taxes on crops and other goods. The import tax for commodityinputs as fertilizer, .;c-;ds, and ago-chemicals was also removed. This may have contributed to the increased use of f~riilizerthat has been observed in the past decade (Pender et al., 2001), and influenced the :xport volun~esof coffee, cotton, and other crops. The exchange rate liberalization is likely to have increased the competitivenessof export crops. As part of the financial sector liberalization, ~icrc2-financeinstitutions were allowed to operate; hence, increasing credit access among the rllral fidrmers. These international trade and financial sector reforms are likely to have affected l a d management,but it is extremelydifficult to assess the direct impacts. '"3. Market liberalization and privatization. Agricultural reforms were geared toward enhancing growth of the private sector and reducing the role of the public sector in production and marketing. One major institutionalreform that is likely to have 2ffected land management is the divesture of total abolition of crop marketing boards. Consequently, the participation of the private sector in agricultural input and output trading increased significantly. Due to market liberalization, the farmers' share of the international prices of major traditional export crops increased from 30 percent to about 79 percent (Balihuta and Sen, 2001). Undoubtedly, these changes impacted land management, because they affect production incentives. However, given the interplay with other important factors, it is not possible to qualify these linkages further. 74. The analysis of proximate and underlying causes of land degradation in Uganda clearly reveals that the causes are multi-faceted, complex and context-specific. Considered in isolation and countrywide each factor is ambiguous with no clear-cut and definite answer to rhe land degradation issue. The causes may vary depending on the nature of the specific underlying socio-economic factor as well as the type of land degradation or type of SLM considered. The interplay of various socio-economic factors needs to be aclcnowledged while designing SLM strategiesand allocating public resources for a specificregion. 75. Another important preliminzry conclusion is that land degradation and sustainable land management are deeply embedded in Uganda's socio-economic development context. Degradation of the land resourc: base undermines the ability of land use sectors (in particular agiculture and forestry) to unleash their full potential as a development tool and contribute to cconomic growth, poverty reduction and environmental sustainability. Sustainable land ~~anagementis not a pure environmental issue, but a key element of the agriculture-for- Jevelopment agenda. SLM sector policy reforms will be discussed in section 3.3. We do not discuss decentralizationfilrther, as it has been already discussedabove. 3. SLM-RELATED INSTITUTIONS, POLICIES AND STRATEGIES 76. This chapter begins with defining a country-specific concept of SLM for Uganda. After that the institutional landscape for SLM in Uganda will be discussed,covering local, national and international levels. The SLM sector strategies and investment plans are analyzed, with the aim of determining how well sector policies are linked to the budget planning process and whether articulated sector priorities are reflected in the actual spending. Finally, the chapter identifies different institutional bodies and mechanisms that exist for collaboration among different actors and discussesthe challengesthey confront. 3.1 Definition of Sustainable Land Management in Uganda 77. Many definitions for SLM exist. Their variety reflects the complexity of socio-economic and environmental relationships involved. Environmental characteristics, market forces, social ambitions, development objectivesand conservation aims are examples of the forces and factors that interact to determine sustainability. A SLM definition for public expenditure reviews should provide a basis of identifying: (i) the institutional landscape for SLM; (ii) the relevant policies, strategiesand investment plans; (iii) the SLM portfolio for the analysis of public expenditure. 78. For the purpose of the PER, the SLM definition needs to be operationalized and made country-specific.According to TerrAfiica (2005), SLM is defined as "adoption of land systems that, through appropriate management practices, enables land users to maximize the economic and social benefits from the land while maintaining or enhancing the ecological support functions of the land resources". TenAfrica (2005) suggests to operationalizethis definition and make it country-specificbased on the followingprinciples: Keep thefocus of the SLM definition on the maintenance of the productive potential of land resources which places it firmly in the economic growth area rather than in the ecological/biodiversity conservation area and on land rather than on the environment in general. Choose tofocus the definition of SLM on those sectorsfor which land sustainability is a critical issue (usually agriculture and forestry). Actions on SLM in these sectors will automatically contribute to the improvement of conditions of sectors suffering indirectly from land degradation. 79. Based on the conclusions of the previous sections that agriculture and forestry are the most important land use sectors in Uganda with regard to socio-economic development and given the above criteria, this PER focuses mainly on policies, strategies and expenditure of these two sectors. However, with regard to the institutional landscape a broader overview is provided. TerrAfiica refers to agriculture and forestry as "SLM sectors" and other relevant sectors as "SLM-related sectors". 3.2 TheInstitutional Landscapefor Sustainable Land Management 80. Since SLM is a cross-sectoral issue, a wide range of institutions are involved in the relevant budget processes (or influencing them in some way) and in preparing and implementing SLM-related activities. Figure 4 provides an overview of the different organizations at the national, international, local government and community level (for more details on the budget process see chapter4).7 3.2.1 Organizations at the National Level 81. SLM activities mainly fall in the mandates of four different Ministries, which are in charge of several semi-autonomous agencies that are also involved in SLM-related activities. The Ministry of Agriculture, Animal Industry and Fisheries (MAAIF) is in charge of promoting sustainable agriculturalland management activities, sustainablerangeland and pasture management activities, and sustainable fisheries activities. It initiates and formulates relevant sustainable agricultural land use policies, standards and guidelines and provides technical assistance to districts. The MAAIF houses the focal point for the United Nations Convention to Combat Desertification (UNCCD) and reports to the Conference of the Parties. It liaises with other UNCCD National Action Plan (NAP) implementing institutionsand programs The F m . Development Department of MAAIF also deals with watershed management, including soil and water conservation and irrigation and drainage. MAAIF has the overall responsibility for NAADS, which was established in 2001 as a semi-autonomousbody under MAAIF. NAADS has a strategy that aims at integrating natural resource management issues into NAADS work plans and into the contractswith serviceproviders. The Ministry of Water and Environment (MWE) is responsible for sustainable forest and woodland management activities and for integrated watershed management.9 Development of water resources for production also falls under the mandate of this Ministry. The Ministry oversees the National Forest Authority (NFA), which is in charge of managing the country's Central Forest Reserves in a sustainable way. The NFA also provides policy guidelines for collaborative forest management and sustainable energy production. In addition, the Ministry oversees the National Environment Management Authority (NEMA), which has the mandate to monitor, plan and coordinate environmental matters, while the implementation remains the responsibility of the relevant line ministries. The Ministry of LanL, Housing and Urban Development (MLHUD) has the responsibility for ensuring"security of land tenure and productive use of land resources." The Ministry is in charge of land administration, which includes land registration and the development of a Land Information System. It has been in charge of developing a National Land Policy, a draft of which was presented to the public in February 2007. 'Due to space limitations,not all linkages between the organizations could be presented in Figure 4. Linkages and coordination mechanisms between different organizations and levels are, however, fhther discussed in section 3.4 and Annex 3. Uganda signed and ratifiedthe UnitedNations Conventionto Combat Desertification(UNCCD) in 1994 and 1997, respectively. Since 2007, the Ministry is called the Minisby of Water and Environment. 'Land' was transferred to the Ministry of Lands, Housing and UrbanDevelopment. The Ministry of Tourism, Trade and Industry (MTTI) oversees the Uganda Wildlife Authority (UWA), which is in charge of managing the country's protected areas. To protect wildliferesources, UWA also works with communities outside protected areas. The Ministry of Energy and Mineral Development (MEMD) is in charge of energyrelated SLM activities, such as the promotion of renewable energies and rural electrification. 82. Apart from these line Ministries, the Ministry of Finance, Planning and Economic Development (MFPED) plays a central role in the budget process, obviously also affecting resource allocation for SLM-related activities. The MFPED also houses the Global Environment Facility (GEF) national focal point. The Parliament plays a key role in the budget process as well and has committees that correspond to the Ministries. Hence, there is a Natural Resources Committee arid an Agricultural Committee. For the purpose of the budget, sectors have been defined, which may cut across the portfolio of different Ministries, or cover only parts of the portfolio of a Ministry. The sectors that are relevant for SLM are the agricultural sector, the environment and natural resources sector, the lands and housing sector, and the energy sector. In each sector, there are Sector Working Groups, which are typically led by the Permanent Secretary of the lead Ministry, and which include members of the civil society. As further detailedbelow, they play a dominant role in the budget process. 83. Further civil society organizations that work on SLM are involved in budget processes. NAADS has a national Farmers' Forum, and there are several independentfarmers ' organizations, which have formed an umbrella organization, the Uganda National Farmers Federation. Likewise, there is a wide range of NGOs, includingthose focusing on environmental issues, and those focusing on rural and agricultural development.NGOs involved in sustainable land management include, for example, the Uganda Land Alliance, Environment Alert and Ecotnist. A range of private sector enterprises is also of relevance for SLM activities. They include enterprises involved in forest plantations, fisheries, land valuation, tourism and coi~sultancyservices. 84. Agricultural and environmental research organizations and think tanks are also important stakeholder with regard to SLM. They include the National Agricultural Research Organization (NARO) as well as University-based research institutes, such as the Makerere liniversity Institute of Environment and Natural Resources. 3-22.rjlfernational Organizations, Programs, and Parinerships - - a:, %$ ec.ali kternational organizations, programs, and partnerships play a role in SLM. t;q.>::n- 2:gani~ationsand intelmationalfinancial institutions, such as the World Bank,finance 2 considerab!e share of SL,M activities, as shown in this report. They also provide technical expertise. Numerous multilateral organizations such as the African Development Bank (AfDB), the Food and Agriculture Organization of the United Nations (FAO), or the United Nations Development Program (UNDP) implement land management operations and engage in policy ti~alogaeon SLM issues. Donor organizations have formed donor coordination groups that are nligled with the sectors. The ones that are most relevant for SL,M include the donor coordination p u p in charge of the agricultural sector, and thdt in charge of the environment and natural lesowws sector. International NGOs also finance SL,M activities and provide technical support. As an example, a number of international conservation NGOs contribute to the financing of protected areas, and hence to UWA's budget. 25. The .Alliance for a Green Revolution in Africa (AGRA) is preparing a major Soil iiralrh Initiative that aims at promoting locally appropriate soil management practices that cnr~hii~t: rise of organic matter and fertilizer to restore soil health. AGRA recently announced he several new programs to be carried with NARO and private seed companies in Uganda. 85. The Government of Uganda, NEPAD, COMESA, and most of the international organizations are already involved in TerrAfrica, a broad SLM partnership between Afiican rau~tries, regional organizations, and international actors. TerrAfrica directly supports mpkrr~entationof the C A M P and UNCCD agendas to improve agricultural productivity on an crir iro;l:gtytltaI!y sustainable basis. TerrAfrica provides a common platform for aligning ir,\lr;slnl,:ni ant! knowledge by supporting program based approaches to scale up SLM, such as is e*qzrr_ir:sit TJgmda. In addition, Uganda is a member of the TerrAfrica Executive Committee ;,rid l i r , ~ -xpr;sseJ its intent to shift to a program based approach to SLM under the CAADP fnalnei$c~r~ bpi!lar 1oil 13116 and water management). ,,-.i hli~~.~I brsvernmenaSystem . y .k,':.;+j?..c . .<. gc4v;:rnrneni system h a five tiers: District, County, Sub-County, ;:-LC. . - . ' tc bijdget processes, the County and the Parish level play a 3 T ; ! ' ! , -. ';1:ii,!fs re-;,..'- . , , + . ! I <> d. ,,. . icp.c.;2: r ? -.. 7 , . ..,";, ,.23*"..:..;lbr:, rFe discuss~onhere concentrates on the District, the Sub-county and the i . '";. 1 .,.., i + ,...j s . .-0 , - . +. !'lli!,*ii8r I. vtd; The public administration at the District level is headed by the Chief , Officer (CAO). Under the District Administration, the Directorate for : _:-.. :.;::;:rai;vc . . . :- - ~. hl:ukelli~g and A-gricultural Extension Services is in charge of agricultural . . tioi.; 3,.tilc;s, i CJirecto ate also oversees the Distiict NAADS coordinator. The District i.;!:,":?r: y Services manage the forest resources in the District (except those managed by i..:i.ii,i+rldLiWA), and ihe Department of Larid Management is in charge of environment .,aidwztj ':k:.= L>istnct 'I'ccitrical Planning Committee (DTCE) plays a central role for the budget .,,* , . ..-...,,.= . *,. ., 'The political head of the District is the District Chair, the relevant political body .. .:..$ , . !- i_!.:lrr!-! {ZGU::!:.:~,~.vhich,similarly to the Parliament, has a committee r:tructure. The District Executive Committee consists of the District Chair and the District Secretaries, who may be considered as "Ministers" at the District level. Sub-county level: The sub-countiesdevelop and execute their own budget. Similar to the District level, they have administrative staff, a Sub-county Executive as political head, and a Sub-county Council of elected members. NAADS awards contracts to service providers at the sub-countylevel. Village level: Village councils and village assemblies are supposed to play a key role in the envisaged bottom-up planning and priority-setting. 89. Next to the government organizations, typically several non-governmental organizations operate at the local government level. Under NAADS, farmers' groups are formed at the village level. At the Sub-county level, a Sub-county Farmers'Forum is formed, consisting of six members. Three members form the procurement committee, in which the sub- county NAADS coordinator and the Secretary for Production are also involved. The other three members form the Executive Committee. The guidelines prescribe that there should be at least one woman in each of the two committees. At the District level, a NAADS Farmer Fonun is formed, as well. SASAKAWA 2000 is an example of an NGO that promotes sustainable land management and operates at the local level. At the village level, there maybe other groups besides the NAADS farmers' groups, which are involved in SLM. Examples are forest user groups, or groupsthat deal with irrigation. Figure 4: Institutional SLM landscape Legend Farmers' Organizations/Village- Obased organizations I 1 Elected 1political bodies National Ievel President (Publicadministration-agriculture - b e Minister Public administrahon: environnent Public administration- general Plan for the Ministry of Finance, Non-governmentalorganizations Research Private sector Modernization of . Planning and Economic organizations Private sector enterprises enterprises Agriculture (PMPc) Development(WPED) and Minerals Agriculture, Animal National Urban Dev. (MAAIF) Ministry of Tourism Trade and Industries National NationaiForest Environment Uganda Authority(NFA) Wildlife AgencyWEMA) Secretariat Authority ChiefAdmkxistrative District District Councils Local government -- OPticer(CAO) - Chair Comrnjttees Ievel Production NAADS District farmer fora . . . . . . . . . County level instltuhons i . . . - - Sub-county Sub-county Sub-county Sub-county Councils - staff Executive Committees ,............................................................................... i- Parishlevel institutions : ............................................................................. I 7 I Resourceuser 22 Other village-based Village Councils Communitylevel SOUPS organizations 3.3 SLM Policies and Strategies 90. This section discusses the policies and strategies of "SLM sectors". In particular, it will be analyzed to which extent SEM issues have been identified as central pillars of the respective strategies and to which extent this "green issue" of SEM is seen as an integral part of the overall economic growth and poverty reduction agenda. The SEM-related priority areas for action will be identified together with planned budgets to allow for an analysis of the alignment with expenditure in the following chapter. Furthermore, it will be discussed to which extent the strategies provide a rationale for public investments and whether the exact role of the public sector (i.e. outright provision, financingor regulation) has been identified. 3.3.1 Agricultural Sector Plans 91. The Plan for Modernization of Agriculture explicitly mentions sustainable management of natural resources in its vision, mission, and objectives. The Plan .for Modernization of Agriculture (PMA) was prepared as part of the Poverty Eradication Action Plan (PEAP) in 2000. The PMA's vision is "poverty eradication through a profitable, competitive, sustainable and dynamic agricultural and agro-industrial sector". Vision, mission, objectives and priority areas for action all explicitly mention sustainable management of natural resources as key to the development of the agricultural sector. The PMA is envisioning an agricultural sector that is "effectively utilizing resources such as land, water and forests in a sustainable manner for both the present and future generations." The objectives include to '>remote sustainable use and management of natural resources by developing a land use and managementpolicy and promotion of environmentallyfriendly technologies". 92. The PMA acknowledges environmental degradation, in particular land degradation as one major agricultural sector constraint which is leading to low productivity. The plan reports that farmers identified soil infertility and soil erosion and deforestation as major concerns. Further, the plan identifies the development of technologies for soil fertility management, water conservatioil and environmental protection as one of the main thrusts for public action in the modernization of agriculture. 93. The PMA explicitly states that SLM issues and concerns would be incorporated in all its six programs. All six priority areas for action of the PMA explicitly incorporate SLM-related concerns and actions. The first priority area "research and technology development" identifies technology development for land and water resources as a focal area: "Soilproductivitypresents one of the major challenges to agricultural modernization in Uganda. Therefore, research will also put emphasis on arresting soil degradation arising from current practices and iciextzfy practices and technologies that will restore and increase soilfertility in a sustainable manner." The "national agricultural advisory service" - the second priority area - emphasizes the importance of advising farmers on environmental management. Managing natural resource productivity is one of the guiding principles of NAADS and is incorporated into N M S ' work plans. The "agricultural education" priority area mentions the importance of including environmental aspects in curricula of relevant educational institutions. The "improving access to rural finance" priority area acknowledges the importanceof access to loans for purchasing inputs and making long-term investments - both investments are crucial for sustainable land management. The fifth priority area "agro-processing and marketing" highlights that declining soil fertility is a consequence of limited fertilizer use (which in turn needs to be addressed through better market access).Finally, "sustainable natural resource utilization and management" constitute another priority area for action under the PMA. This priority area highlights the importance of technology development for SLM and land reforms as a necessary condition for sustainablemanagement of natural resources. 94. In general, the PMA defines clearly the roles of the public vs. the private sector. The plan restricts the role of the Government to the provision of services which are "public good" in nature. The PMA aims at removing direct Government involvement in commercial aspects of agriculture and at promoting the role of the private sector. The Government is envisaged to create an enabling environment for the private sector through policy development and strategic planning; provision of a conducive legal and regulatory framework; setting and enforcement of standards and regulations; information and data collection and dissemination; capacity building of various stakeholders in the agricultural sector (such as service providers and farmer organizations). 95. Furthermore, the Government would promote the private sector by expanding opportunities for the private sector to participate in policy formulation processes and implementation of publicly funded programs. The private sector would be empowered through their involvement in sector plans and by being contracted for direct delivery of public services to farmers on a commercialbasis. Opportunitieswould be explored for the involvement of the private sector in research and delivery of advisoryservices. 96. The PMA acknowledges the importance of decentralization for an efficient and effective implementation of the plan: "The PMA will be implemented through decentralized planningprocesses which will identlfi the key constraintsat the local Government level and help to remove these constraints through 'yoined-up"public sector interventions involving policy adjustments when required or public sector investments or new public service delivery mechanisms as will be found to be appropriate." According to the PMA the functions which have been decentralized for implementation by the Local Government include: (i) control of erosion, bush fires, local hunting and fishing; (ii) management of forests and wetlands; (iii) land administration and surveying; and (iv) agricultural extension for crops, animals and fisheries. Decentralizationis particularly crucial for SLM, since it can be very much considered as being a location-specific issue which is difficultto be addressed at the national level only. 97. The PMA emphasizes multi-sectoral approaches across Ministries and agencies for the implementation of the plan. The PMA highlights the impoitance of NEMA: "Emphasiswill beput on ensuring eflective linkages between the Agricultural Advisory Services, the Production and Environment Committees at various local government levels, and the environmental awareness activities of NEMA to influence the attitudes of those presently contributing to environmentaldegradation." 98. The Development Strategy and Investment Plan 2005-2008 (MDS&IP) by the MAAIF confirms the PMA's focus on sustainable management of natural resources. The MDS&IP aims at translating the goals and priorities indicated in the PMA into a specific plan for public sector activities, including public spending. The MDS&IP confirms cardinal principles underlying the PMA, such as decentralization,empowerment of local governments, limiting the role of the public sector to provision of core public goods, and coordination of multi-sectoral interventions. 99. The MDS&IP highlights the need for improved SLM technologies. It underlines that soil erosion and nutrient depletion are serious problems which are negatively affecting agricultural productivity. Therefore, the plan includes sustainable land use and management as one of the sector's major objectives and expresses "the needfor continued emphasis onfertility- replenishing inputs and improvedplanting materials, the adoption of soil conservationpractices and the use of water harvesting technologies". The plan highlights the fact that all MAAIF activities are in line with the objectives of the National Environment Management Policy (1994) of promoting farming systems and land use practices that conserve and enhance land productivity in an environmentallysustainablemanner. 100. The MDS&IP points out that environmental issues are cross-cutting and focuses two of its intervention areas directly on SLM. The MDS&IP identifies 13 intervention areas for which it identifies policy objectives,justification for public involvement, outputs, activities, and budgets. Environmental issues are identified as being cross-cutting and considered in each intervention area. However, two intervention areas explicitly focus on SLM, namely "capacity building for sustainable soil and water utilization, agricultural mechanization, and rangeland management" and "promotion of increased agricultural production and productivity". The justification for both areas is based on the public good argument and externalities.In general, the role of the public sector is restricted to regulation, capacity building and financing rather than outright provision of goods and services. The policy objectives of the former intervention area includes capacity building of the farming population and extension service providers, promote the use of appropriate technologies, develop standards for sustainable management of natural resources, and monitoring and evaluation. The total allocation for this intervention area is UGX 55.44 billion in 2005-2008. The majority of the finds are allocated to the promotion of . appropriate technologies and practices (UGX 31.2 billion), followed by irrigation and water harvesting (UGX 19.8 billion), and capacity building of MAAIF, local government and service providers (UGX 1.3 billion). The policy objectives of the intervention area "promotion of increased agricultural production and productivity" go beyond pure focus on sustainable management of natural resources, but include the promotion and support of sustainable utilization of farmland resources for increased productivity and protection of the environment. The total allocation for this area is UGX 2.54 billion, whereas UGX 1.2 billion are allocated for "promotion of strategic commodities and technologies" and UGX 0.8 billion for "capacity building of service providers". 3.3.2 Forestry Sector Plans 101. The Uganda Forestry Policy envisions harmonization of economic and environmental objectives on both government and private land. The Ugandan Forestry Policy (MLWE, 2001) provides clear directions for the development of the forest sector. The policy is cone-+-* tJ). I f 8 'I'he principles of the ENR SIP implementation strategy include decentralization, private sector participation, and inter-sectoral coordination. However, concrete activities on I:( 3~ ic: rneet the objectives and targets are not included. Consequently, it is not entirely clear hrilat the role of the public sector vis-a-vis the private sector would be. While alternative fi~~ancmgis envisaged (e.g. private investments, payment for environmental services, carbon iil:an%::>),tit:: major part of the budget is expected to be financed by the central and local gc *~r:~~~ments,and bilateral and multilateral donors. .-I. '-4 0thr.r SLM-relevrrnt sectorplans 0 Other SLM-relevant sector plans include the Land Sector Strategic Plan, the N,~tia)malBiodiversity Plan, the Uganda Wildlife Policy, the Wetlands Strategic Plan, and the 'TourismPolicy for Uganda. Since this SLM PER focuses on the "SLM sector" - agriculture ard forestry - a detailed discussion of these plans is not provided. However, in general all these se-atorplans emphasize the civcial role of sustainablemanagement of land resource for economic gr jwth and poverty reduction and are based on the guiding principles of decentralization,private sector qarticipation, community involvement and multi-sectoral approaches. i i ). Uganda's National Adaptation Programs of Action (Republic of Uganda, 2007) places tht issue of land degradation firmly in the context of climate change. Climate change is cc;asitf~:redas a cause of land degradation. At the same time sustainable land management uyrior~sare identified as suitable adaptation options. 111 As a conclusion, SLM-sector strategies and investment plans provide an economic ra(i.ionaalfor public investments in sustainable land management and clearly define the role af the private sector vs. the public sector. Land degradation is acknowledged having not only en ~iror~menta!!y,but also severe socio-economic consequences. SLM is considered to be crucial to reach economic growth and poverty reduction targets. Strategies and investment plans integrate SLM systematically in their objectives and central pillars. The private sector is expected to make investments in profit-generating private goods, such as in inputs needed to adopt and maintain financially attractive SLM practices. The public sector's role is to create an environment that favors these private investments, for example by making investmentsin SLM- related research and development, and by providing adequate institutional and physical infrastructure, an adequate regulatory framework, capacity building and education, and knowledge management. 3.4 Coordination Mechanismsfor SLM: Opportunitiesand Challenges 112. This section identifies different institutional bodies and mechanisms that exist for collaborationamong differentactors and discussesthe challenges they confront. 3.4.1 Institutional bodiesfor SLM coordination Plan for the Modernization of Agriculture (PMA): The PMA is a rather unique cross- sectoral coordination mechanism for agricultural development. The PMA addresses the challenge that promoting agricultural development requires the coordinated action of several Ministries besides the MAAIF as well as collaborationwith the private sector and civil society. The institutional structure of the PMA includes a multi-stakeholder Steering Committee of approximately 20 members from the public sector, the private sector and civil society, a Forum with an even broader membership, and a Secretariat,which is part of the government structure under MAAIF. The PMA pillars are guided by committees that are supposed to ensure coordination between the different organizations involved in the respective pillar. With regard to the Natural Resources Pillar, the PMA stresses the coordinationwith NAADS and with NEMA. Sector Working Groups: Since the budget process is based on sectors rather than individual Ministries, there are established coordination mechanisms within sectors. The Sector Working Groups offer the possibilities to coordinate between Ministries and with civil society and the private sector. In particular, the Sector Working Group for Environment and Natural Resources offers good possibilities for coordination regarding SLM. However, SLM cuts across different sectors, especially the Environment and Natural Resources Sector, the Agricultural Sector, and the Energy Sector. According to the interviews with representatives of MFPED, the composition of the Sector Working Groups should not only ensure that all relevant organizations within the sector are represented, but also include representatives from organizations that are important for coordination with other sectors. The major challenge faced by Sector Working Groups seems to be their mode of operation, such as the frequency of their meetings, rather than their structure. Donor working groups: Since donors fimd a large proportion of SLM activities, donor coordination is also very important in this field. Donors and international financial institutions have formed coordination groups that correspond to the sectors. Since SLM cuts across several sectors, it also cuts across several donor coordination groups (e.g. agriculture; environment and natural resources; energy). Since the topic is relatively new on the agenda, it was still unclear how the different coordination groups will deal with this cross-cutting issue. The National Environment Management Authority (NEMA):NEMA was created with the goal to facilitate coordination across Ministries and Agencies on environmental matters. NEMA has a Policy, Planning and Information Department that has the explicit task to ensure collaborationwith all Ministries and Agencies concerned. Likewise, NEMA has a District Support Coordination and Public Education Department that should facilitate coordination with the Districts. The capacity of NEMA and the appropriateness of the institutional framework under which it operates are widely acknowledged. The major constraint seems to be the ability to enforce environmental regulations on the ground. Major responsibilitiesfor environmentalenforcement rest with the Environment Officers, who are staff of the Districts, not of NEMA. A study on decentralized natural resource management showed that local governments often lack the capacity to exercise the authority they have with regard to natural resource management. The lack of environmental officers at the Sub-county level was noted as one constraint (Bazaars, 2003). SLM Technical Working Group and Steering Committee: An "Inter-ministerial Framework for Cooperation on the Development and Implementation of a Country Program on Sustainable Land Management in Uganda" has been developed by MAAIF, MWE, MEMD, and MLHUD. This Framework identifies MAAIF as the Focal Point Ministry, which is in charge of ensuring appropriate Secretariat functions for the collaboration. Two committees have been established under this framework, a Technical Working Committee, and a multi-sectoral Steering Committee as the main decision- making body in the partnership to be co-chaired at the Permanent Secretary level. Since this structurehas been created recently, its functioningcould not yet be assessed. 113. The Inter-ministerial Framework agreement includes a general outline of priority thematic areas for coordinated donor support and clear responsibilities amongst the ministries. The framework agreement specifically remains open to involvement by additional agencies and the districts, which were not represented in the agreement.The agreement builds on the past two years of internal and international dialogue and momentum on land management. This framework agreement is in part a result of the process initiated by the ministries in 2006 to: (i) improve alignment between the agricultural and environmental priorities including the country's UNCCD National Action Plan and CAADP processes and the Millennium Development Goals (MDGs); and (ii) implement these priorities by scaling up SLM. To support this effort, the inter-ministerial agreement includes an explicit pledge by the ministries to collaborate in the developmentof a national SLM InvestmentFramework as a key tool for long- term implementationof the country's nascent SLM Country Program. 3.4.2 Ad hoc mechanismsfor SLM coordination 114. In addition to the institutional bodies that exist to promote collaboration for SLM, respondents pointed to a range of ad-hoc mechanisms for coordination,which can be used to address specific problems, which do not require a permanent institutional set-up. For example, problems that c.lf across Ministries can be addressed at the regular meeting gf all Permanent Secretaries, which takes place every month. Another example is the creation of an inter-ministerial Task Force to address the problem of encroachment of protected areas by pastoralists. This problem had become a political issue, and the President had written to the Ministries in charge to resolve the problem. The high-level attention facilitated the collaboration on this issue. There are also examples of projects that are jointly managed by two Ministries. With regard to the budget process, there is also the possibility that the Working Groups of different sectors hold joint conferences when they are in the phase of preparing their Budget Framework Papers, as was pointed out by the Permanent Secretary of MAAIF. The serni- autonomous agencies UWA, NFA and NEMA reported a good experience of collal~orating informally on various issues, even though they have apparently faced problems with their parent Ministries when trying to formalizepartnerships through Memoranda of Understanding. 115. Communication procedures seem to be a major challenge for inter-ministerial collaboration. If a staff member of one Ministry wants to communicate officially with a staff member of another Ministry, the communication has to be routed through the Directors of the respective Departments and the Permanent Secretaries of each Ministry. Communication channels to Districts are rather challenging, as well. For example, if a staff member of the Ministry of Agriculture wants to correspond with a colleague at the District level, the correspondence has to also be routed through hisher Director in the Ministry, the Permanent Sectary, the Chief Administrative Officer of the District, and the head of the respective Directorate at the District level. By contrast, staff members within the same Ministry can easily communicatewith each other; they only have to copy their respective superiors. 3.4.3 Mainstreaming provisions for SLM coordination 116. A third mechanism for coordination of SLM activities are provisions that require the mainstreamingof environmentalconcerns. Mainstreaming environment in Budget Framework Papers: NEMA has developed Guidelines for "Mainstreaming Environmental Issues into Budget Framework Papers" (NEMA, 2007), which has been distributed by MFPED to the Sector Working Groups. From 2007 onwards, the Guidelines will be issued as an Annex to the Budget Call Circular, thus requiring all Ministries and Agencies to take environmental issues into account when preparing their budgets. Without doubt, this mainstreaming provision is potentially a very powerful tool to address cross-cutting SLM issues in the budget process. But since the mechanism is new, it is not possible to assess its effectiveness yet. According to NEMA, informal communicationwith the different ministries and agencies during the preparation of the Budget Framework Papers has proved to be useful for mainstreaming. At present, MFPED seems to have limited capacity to assess to which extent the Budget Framework Paper of each sector has in fact cori~pliedwit11 tlrc mainstreaming guidelines. Mainstreaming sustainable resource management in NAADS: As indicated above, as part of its strategy for sustainable natural resource management, NAADS has developed guidelines for incorporating natural resource management issues into work plans and into the contracts made with service providers at the sub-county level. In fact, agricultural advisory services need to play a key role in promoting SLM activitiys 01, 111.. g(.11 -:1' Therefore, thc guidellties for mainstreaming SLM issues in work plans :mi(.c)s 4;; #:is c ,' NAADS are potentially a very powerful tool. However, the act~a?~r~pir,r~-cr::- " these guidelines remains weak and SLM issues have not been systematically addressed in NAADS programs. Mainstreaming in agricultural research: "Regard for the protection of the environment" is specified among the objectives of the National Agricultural Research Organization (NARO) in the national Agricultural Research Act of 2005. Although environmental sustainability issues have been mainstreamed into all research activities, on-the-ground dissemination of these techniques remains unsatisfactory. Partially, this can be again explained by the commodity focus of NAADS without necessarily promoting SLM techniques on farm level. a District Environment Action Plans: Another mechanism for mainstreaming SLM activities are the District Environment Action Plans, which have been promoted under donor-funded initiatives, including UNDP. However, as indicated above, ensuring appropriate funding for these plans in the absence of a special program seems to be a major challenge. 3.4.4 How to improve SLM coordination mechanism? 117. Overall, the formal institutional framework and the processes in place for budget planning and execution are well developed; in fact, they may be among the most advanced in Sub-SaharanAfrica. Likewise, there are advanced mechanisms for institutional coordinationand for mainstreaming environmental matters, including SLM, into sector plans and budgets. 118. However, major problems stem from the lack of capacity and incentives to use these mechanisms effectively. Therefore, it can be concluded that, rather than creating new institutional structures and processes for SLM, a more promising strategy would be to address the problems of capacity and incentives. The following recommendations should support this strategy: a Developing a shared understanding of SLM The concept of "Sustainable Land Management", as framed by the TerrAfrica initiative, CAADP, and UNCCD, still seems to be a rather new concept for many actors in Uganda. The current framing of environmental matters has evolved around the concepts of "environmental protection" and "sustainable natural resource management", with a focus on forests, rangelands, wetlands, watersheds and water resources, wildlife and biodiversity. The major contribution of the SLM concept is to add the sustainability of land under agricultural production to this portfolio, or in other words, integrate the SLM into agricultural land productivity agenda. This is an important addition, since the problems of soil degradation and nutrient depletion of agricultural land have so far received limited attention within the environmental agenda. a Integrating SLM inplanning documents: Since the upcoming National Development Plan and the sectoral Development Strategy and Investment Plans form the basis for budget allocations and budget performance reviews, it is essential to ensure that SLM concerns are adequately addressed in these planning documents. If SLM concerns are not among the priority areas in these plans, it will be difficult to ensure an adequate budget allocation for SLM. Joint programs that require a coordinated budget allocation in two different sectors need to be addressed in the phase of strategic planning to facilitate coordination in the budget process later on. Generating data to make the casefor SLM In negotiations regarding strategic priority setting and budget allocation, it is important to show: (i) which interventions work on the ground to solve socio-economic and environmental problems; (ii) how much they cost; and (iii) what the financial and economic rates of return have been. So far, the limited availability of such types of data has been a challenge for ensuring an adequate budget allocation for SLM in the budget process. Even if the urgency of SLM problems can be made clear in budget negotiations, making the case for an increased budget depends on the ability to offer convincing solutions and monitor SLM indicators. At present, it is still difficult to point out which solutions will work for some important SLM problems in Uganda, such as nutrient depletion, what the investment requirements are, and what returns can be expected. The generation of results would need to begin with assessing the impact of past and on-going SLM projects. It is also important to strengthen the analytical capacity to generate the above information, and to evaluate public spending on SLM in terms of returns. Such information will also make it easier to move fiom the current activity- and output-based performance reporting in Budget Framework Papers to an outcome and results-orientedreporting. IdentzJLing areas where coordination is essential, but avoiding too much coordination: While coordination is obviously important for SLM, coordination involves costs and is not a goal in itself. With regard to the SLM agenda, it will be important to identify those activities that clearly fall under the mandate of one Ministry and require limited coordination with others for their effective implementation. For example, creating incentives for f m e r s to reduce nutrient depletion and taking measures to reduce soil erosion clearly fall under the mandate of MAAIF. For the sustainable development of water resources, however, collaboration between MAAIF and MLWE is essential. Likewise coordination between MAAIF and UWA and NFA is important to address encroachment problems. The types of coordination mechanisms that are most suitable may differ depending on the problem to be solved. Identifying already existing coordination mechanisms that can be used may be preferable to the creation of new mechanisms, because every new mechanism, such as a new committee, will compete for scarce staff time and resources within all agencies concerned. Likewise, for SLM, using the existing mainstreaming provisions effectively (see above) appears more promising than creating new mechanisms. Making communication procedures easier between technical staff of different Ministries, and between technical staff of Ministries and Districts, would also help to facilitate coordination. Strengthening the role of SLM-related civil society organizations in the budget process: As has been highlighted above, Uganda's budget process provides considerable scope for the involvement of civil society. As the experience world-wide has shown, civil society participation can play a key role in bringing sustainability issues on to the political agenda. Yet, civil society organizations in Uganda are concerned about the problem of "token participation." To address this problem, it may be usehl to clarify the role of civil society participation in different processes, and to improve accountability mechanisms for participation at the same time. Participation can be used to provide citizens "with a voice, but not a vote", or to provide them "with a vote." The latter case applies if civil society organizations are represented on governing bodies of public institutions, as in case of NAADS. In the former case, which may be typical for the budget process, not all suggestions made by civil society will be taken up as the ultimate decision rests with political bodies. Clarifying the role of participation and establishing accountability mechanisms will help to make participation more meaningful, even if not all suggestions made by civil society are taken up. This may include (i) distributing invitations and relevant documents (such as budget estimates) early enough to allow participants to get prepared; (ii) prepare reports that take into account suggestions that have been made by participants; and (iii) report back to participants which suggestions have been taken up, and which have not been taken up and why. The role of civil society can also be strengthened for SLM by improving the capacity of organizations to analyze the budget with regard to SLM, and to conduct research on SLM topics that generates useful information for the budget process. As one interviewed Parliamentarian pointed out, civil society organizations have often been most influential when they were able to provide research results that could be quoted in the public debate. In efforts to strengthen the role of civil society participation, special emphasis may be placed on working with farmers' organizations in identifying SLM solutions that work on the ground, and in generating data about such solutions. In view of the challenges of budget execution discussed above, it may also be useful to place more 'emphasis on civil society participation in the monitoring of budget execution, in addition to their participation in budget preparation. 4. IWVIEW OF SLM PUBLIC EXPENDITUIW 119. This chapter analyzes SLM public expenditure.10It builds on the on-going Agriculture PER in Uganda, which analyzes the level and composition of public expenditure for the whole agricultural sector. Since the AgPER describes the budget process at national and local levels, in this SLM PER these processes are summarized in Annex 3. This chapter begins with the identification of the SLM project portfolio and then turns to the analysis of the level of spending and its functional composition during 2001102-2005106. The data is collected across four Ministries, namely Ministry of Agriculture, Animal Industry, and Fisheries; Ministry of Lands, Water and Environment; Ministry of Trade and Tourism Industry, and Ministry of Energy and Mineral Development. The data also includes autonomous public agencies attached to these Ministries, e.g., National AgriculturalResearch Organizationand National Agricultural Advisory Services attached to ILaAAIF; National Forest Authority and National Environmental Management Agency attached to MWE; and Uganda Wildlife Authority attached to MTTI. The analysis of functional composition is based on the reccrded "development capital expenditure"." 4.1 Identification of SLM Portfolio 120. The identification of the SLM portfolio encompasses public expenditure, both government and donor funded. It is based on the standardized Project Information Profiles from MFPED, which include a brief project description, its objectives, expected outputs~outcomes,relevant activities, implementing agency, location, source of financing, amount of planned funds, and some other information. Overall, the identification of the portfolio for SLM PER is a much more complex process compared to single sector PERs, such as an agriculture PER. SLM portfolio identification follows "Financing Guidelines of TerrAfrica" (2007). This is because SLM needs to be defined first and data need to be collected across sectors. In addition to the standalone SLM projectslexpenditure, SL,M components in the agriculture and forestry-relatedprojects are also to be accounted for. Often there was insufficient information about these SLM-related components, which constrained the identification of the SLM expenditure. The SLM portfolio was constructed based on a grouping into "SLM sector", "SLM related sector", and "potential SLM" expenditure.The list of all expenditurelprojectsin all three sub-categoriesis presented in Annex 1. 121 The SLM portfolio does not include off-budget expenditure. Off-budget spending, not recorded in the official budgets, is likely to be significant. According to the Agriculture PER (2008), 20-30 percent of total spending on the agriculture sector has not flowed through either national or local budgets in recent years. Overall, there is little, if any, coherent and consistent information about the donor-financed projects that are off-budget, which requires the attention fiom MFPED to be addressed. 'ODue to limited data availability, this review does not consider private investment in SLM activities. " However, not all development expenditure is capital. All project receipts, revenue as well as appropriations-in-aid from development partners are recorded as development expenditure. Some of these project receipts are recurrent spending such as salaries of project staff, consultancy services, and so forth. According to Agriculture Public Expenditure Review, recently undertaken for Uganda by the Vcrlt! Fa.::. ?he share .-*frrz~lxentexpenditure :n capital development budget can reach 30 percent (World Bank, 2008). 122. Recurrent expenditure was not included in the analysis. This is because a ministry or unit implementing SLM-focused or related activities generally also undertook many other non- SLM related activities. Since no specific budget coding is introduced for SLM in Uganda,joint- cost allocation problems make it impossible to estimate recurrent costs (staff wages and operating expenses) attributable to SLM activities. Therefore, the report does not conduct the economic classification of SLM expenditure. 123. The decision tree for identifying the SLM expenditure portfolio includes several steps. These steps include the selection of a SLM definition, an identification of sectors for which land sustainability is a critical issue, and a consideration of criteria to select SLM expenditure within the identified sectors (Figure 5). The decision tree process is summarized below: Step 1: The identification of the SLM portfolio begins with defining the concept of sustainable land management. As outlined in section 3.1, for this analysis SLM is defined as "the adoption of land systems that, through appropriate management practices, enables land users to maximize the economic and social benefits from the land while maintaining or enhancing the ecological support functions of the land resources". In order to narrow down the analysis to a meaningful and manageable scope, it is necessary to keep the SLM focus on maintenance of the productive potential of land resources. This places SLM firmly in the economic growth area rather than in the ecological/biodiversityconservation area and on land rather than on the environment in general (TerrAfrica, 2007). Step 2: The next step is to ask the question 'tfor which sectors is land sustainability a critical issue"? The sectors, where SLM will automatically contribute to the improvement of productive potential and economic growth, should be grouped into the category "SLM sector". In the Ugandan case, these are the agriculture and forestry sectors. The sectors, which are affected by SLM actions in agriculture and forestry, or which may produce positive and/or negative external effects on SLM sectors, are grouped into the category "SLM related sectors". These can be water, fishery, tourism, protected areas and hydro and energy sectors. Step 3: m a t is a criterion to identzfi SLMprojects within the selected SLM sectors and SLM related sectors? Only those operations were considered which specifically aim at protecting the potential of natural resources and prevent land degradation, and thus sustain the productivity capacity of land in the longer term. Conversely, any operation that potentially impacts upon the productivity of land as a by-product of some other goal, such as increasing agriculturalproductivity, would not be considered as SLM. Figure5: Decision tree for the identificationof the SLM expenditureportfolio Source:Own presentation. 124. Thus, "SLM sector" expenditure in Uganda comprises agriculture (including livestock management), and forestry sectors. For agriculture, SLM is the maintenance of soil productivity over time, combining soil fertility treatment with soil water conservation measures. SLM will focus more on one than the other element of this combinationdepending on the terrain, ecosystems, climate and use of land. For forestry, SLM is the maintenance of the country's forest resources over time. It entails the replanting after harvesting, the protection of biodiversity of forestry resources and the prevention of depletion and destruction of forests. Measures to ensure SLM for forests include the creation of forest reserves with managed access, the prevention of bush fires, tree planting schemes for firewood, logging and harvesting permits accompanied by replanting, and forest management regulations and legislation. For both sectors, SLM includes any activity promoting or ensuring the adoption of SLM practices and technologies by the land users including monitoring and controlling SLM interventions such as user training, advisory services and enforcement of regulations (TerrAfiica, 2007). As a result, the "SLM sector" expenditure is included into the portfolio if its objective is explicitly stated as to support, promote, monitor or enforce sustainable land management or to prevent andlor reverse land degradation or forest depletion; even when these projects have components which are not specifically SLM sector focused, the whole project ought to be considered as a "SLM sector" expenditure and included in the portfolio. 125. In addition, "potential SEI&1["rapesadifuri: was also rdsntdf'Sed. ..,enditare 'jn::ii, ..; is mainly related to overall agricultural prodi.~ctivjtypro_iectswithout 2x1 explicit SLK focus. They encompass the projects that are neither included in the "SLM sector portfolir;" n:r i;l the "SLM related sector portfolio", but could have 5een wed potentiall?, .to include 2I.M components. While standalone SLM projects are cccrssary i s some c.ascsl e.s. to de:jig~SLM-related regulations, the wide-scale adoptioil ijl KZh: ';c;i:,';;;~i;: at fai";~;fc:-t;! :.::suirz:: e-1: !,r;tegra!ion cf SLM agenda into agricultural prodriclivlty agenda in Ugancla. T'k!: ::)t!:;:tihl::e;r:;:-y ;:,~intsc:ar, be providing incentives for SLM adoptio:,, !:j:;17,!cj~rlg szyport. $0 Jf:sjt;*;1 ;. ~17~.;..'\~.a-~7.--'!-- :: ,, l , , . ..ntal srnfices payments, targeted matching grmls or t:~L:ditpri)gi.a;~is,and iinpi-.>vingthe t;a.n~i:erishar~ng of knowledge related to SLM between NA:j.i2S a11.iNARO, and other instiiuiicw:~Krgardi::.lg the NAADS work program, the integration 21- SLM i i ~ advisory senlces oxi prstlu~;ii(~r.tcck~dques should become an effective and efficient way of helping farmers to get acquainted with the results of SLM research, understand the reasons and outcomes 01lalid degaciatioii, and lean1 cost-effective land management practices. 4.2 Trends in the Level of SLM Expenddta~re 126. The actual budget expenditure allocated for "SLM sector9'activities dayring 20!)1/02- 2005/06 was UGX 70.6 billion, or US%38.9 million (Table 1 2nd Figurc 6'). 'Pi;::was about UGX 14.1billion (US$ 7.8 million) per year or G.28 percent of total budget expenditure. Adding "SLM related sector" expenditure increases the share of the SLM expenditure in total budget fourth fold, but still left it at a low I .15 percent. For comparison, the ~ublicspending on the agriculture sector as a share of total budget was 4.3 percent during 2001/02-20i).s!06 (World Rank, 2008). The spending on SLM activities was ::!so low with respect to total G T P . In mite of the fact that agriculture and forestry sertiars creetc:d about 42 percent of Trgands's GDP, the SLM public expenditure on agriculture ar,d fbrestry equaled only !).I3 percent planned of expenditure. According to this Investment Plan, the Mifiistr! sot~ght'tcj all~czit~e UGX 55.4 billion on SLM related activities during 2005-2008. Vv'e use this infom!al,Br~nas a proxy of government plans of how to address land degradation issues and what types nt' speciik SLM activities to finance. 140. After having grouped the MAAIF'q allun:atti?n pPams ;rr,r-ttsn 4'515' eompcak'ieuts, the conclusion can be made that the priority is gJvei~To urs-hf6e-gruunr,b BC~!!.$$:~E? for scualing up SLM (Figure 9). The expenditure for these wtivities :~cco\;:.i:-. :!.:. 75.3 percerlr of total expenditure. The second priority is support of SLh4 re5carr.h acli\<:r :.s !1'7.3 isr?:cer_t).''Together these two priorities (on-the-ground activities :idresearcl:) :-:.scot!:,%??ir?E:S.:i p--rcer~tof'total SLM expendhe, leaving only a small room for spending on enabliil~envir-or;met;r and developing . . SLM knowledge management, M&E, and infornlation :Jis.-;er:~!i1i.!.!{aii!.;l';itk~s.Yo expenditure is allocated to advisory services (NAADS) to suppocr S 1 1 iid..i:?ticb-: t',.ylsrl& !i.ansf'eir!ng the . results of SLM-related research to farmers. ?\i.A.ADSr:; 'iilpr,:.~:.;':j. . I S : : ' t .:q1 ats ^ . V X ' ' , ; . i i g 3 f c~ I . . L Lir, work plans, but in reality this is not done syste,maticallvas di:;c~.ri-i:~:.~ p~'i~.io~~s' !:I :.t:c ri:i:tic.m. . . ' 141. Was the actual expenditure well aligned wit!! the j,bFeb,rra!es ;ss exgrzssed in the investment plans? The analysis of SI,M actual expcr:ditur: by ;~..i::.All d i i n i ~ ~ i,:OCi?/02-2005/06 shows the mismatch between planned and actuai cc;'nl~sosicio~~expeniIi!ur:-::(Figure 'The .. - actual support to on-the-ground activities lror scaling l,ly Sr",147v~.;F. !:;;I rr:a;~zzu as planned, and most expenditure were allocated to research - -d6~0lij:'!i pt:!.ctx;t :.:' ii:?.;~! FJ.l...>d zxpendi[iire. The share of on-the-ground activities reached or;ly i:i perces~:. T2i- psi:riit c:l~- zsperiditure was . allocated to knowledge management, M&:C:., ancj. i~:f~!-I~;.ii!iila;r , :ii?:.~:.:;:?:.:rl.!~!.::: :r: sp;t!3 of' [he Fact . . . that little knowledge still exists to make the casr :ix i:ca'i:-gi,ip 1$1..'?,; .;::try, iik::; 15UGX 4 billion were added annually tn MAAIF's S1M budget. ilslrg Ia~stu.I.:a: :..1 ' Xl i:;)e~idi:~s 8i.Iresearch 311 agenda during 2001102-2005!06. 16It has to be noted that the time periods used for piam;cd and zctu.al r:xpi-rl..~,~~:r, acrcla! and planned expenditure may also be explained partly through the fact that differerrt tilnt. !;150) L_1NOData Source: Own estimate. 150. Overlaying the cost of replenishing depleted nutrients and soil erosion with SLM expenditure indicates good spatial targeting of SLM expenditure in Uganda. 151. Figure 13 and Figure 14 show that SLM expenditure is higher in areas with severe soil nutrient depletion and soil erosion. The Southwestern Highlands, West Nile Area, Lake Victoria Crescent Region, and Southwestern Cattle Comdor show severe nutrient depletion (or the costs of replenishing depleted nutrients) and high targeted SLM expenditure. The exceptions are the Eastern Highlands, the area around Lake Albert (Districts of Kibaale, Hoima, and Bundibugyo), and the District of Luwero. In these areas SLM expenditure was low in spite of severe land degradation. It should also be noted that a couple of districts in Northern and Central Uganda Figure 15: Incidence of poverty and its distribution in Uganda Source:Emwanu et al.(2004). 152. Finally, the spatial analysis was expanded to test overlaps between land degradatiou and poverty hotspots. While the highest incidence of poverty is found in the North and Northeastern Regions, severe land degradation can be observed in the Southwestern Highlands. Lake Victoria Crescent Region, the Southwestern Cattle Corridor, the Northwestern Region and Eastern Highlands. Accordingly, only a few districts in Northwestern and Southwestern Uganda. fall into both categories land degradation hotspots and poverty hotspots. Above, it was indicated that some districts in Northern and Central Uganda received significant SLM resources although land degradation is not severe. A possible explanation might be that the decision of allocating public resources to these areas was more based on poverty aspects than the degree of land degradation. This finding has an important implication for policy-decision making in Uganda- allocating public SLM resources based on poverty hotspots will not automatically reverse severe land degradation. Using public expenditure to promote SLM requires separate targeting. However, expenditure designed to improve SLM can still be pro-poor. Many land degradatosi llotspots are highly populated areas and home to many poor people. This implies that adequate SI,M investments in areas with severe land degradation may lead to growth and poverty reduction. Figure 16: Poverty head count and cost of replenishing depleted nutrients Cost of depleted nutrient rrplenishment & Poverty head count H low cost low powrty low cost high poverty 0medium cost low pouerty medium cost high poverty = high cost low poverty high cost high pouerty 0No data Note: Cost of replenishing depleted nutrients (US$): Low (<1lo), medium (110-150), high (>150); Poverty head count: Low (<40), high (>40). Source: Own estimate. 5. SLM-BASED DEVELOPMENT PATHWAYS FOR LAND DEGRADATION HOTSPOTS IN UGANDA 153. As discussed in chapter 2, the proximate and underlying causes of land degradation in Uganda are complex and context-specific. This implies that addressing these causes and successfully promoting SLM in Uganda through public expenditure, requires a spatially disaggregated and location-specificapproach. Furthermore, chapter 4 points out that SLM public expenditure is relatively well targeted spatially. However, there seems to be a disconnect between SLM sector policies and strategies and SLM programs and projects. While the former emphasizesthe importanceof embeddingconservationof natural resources in the growth agenda of the respective sectors, this is not reflected in the design and implementation of the latter. On the ground activities seem to be isolated from the productivity and growth agenda with a relatively narrow focus on conservationobjectives. 154. The public sector's role is to create an environment that favors these private investments. Examples for public sector roles may include: (i) reducing high up-fiont costs for SLM practices and technologies by generating and disseminatinglow-cost solutions via research and advisory services (NAADS and NARO); (ii) promoting community-driven development approaches based on matching grants (e.g. for development of community nurseries for agro- forestry systems); (iii) providing the regulatory and institutional framework for payment for environmental services systems; and (iv) improving access to in-kind and in-cash credit. However, this general role of the public sector needs to be defined more precisely at a spatially disaggregatedlevel. 155. What are appropriate analytical tools to inform future spatial allocation of SLM expenditure in Uganda? The concept of "development pathways" is especially useful to answer the above questions. This concept is based on the idea that opportunities and constraints for sustainable development depend upon the comparative advantages that exist in a particular location. These comparative advantages determine which type of interventions may yield high social returns. The land management practices that are most profitable and sustainable are also affected by such comparative advantages. Much public action aimed at improving land management focuses on influencing household or community adoption of particular technologies. Yet this may be ineffective if the technologies are not suited to the development pathways that have potential in a given location. It may be more effective in many cases to first focus on the larger development strategies for particular development pathways, before focusing too much on particular land management technologies. 156. The following section will analyze the comparative advantages ("development domains") of the regions covering the identified land degradation hotspots, what trends with regard to land use and land management can be observed, what suitable future SLM-based development pathways may be, and how they could be supported through public intervention^.^' This approach may contribute to the process of prioritizing public investment for SLM in Uganda. The results of the analysis for each development domain are summarized in Table 7. 'OThis chapter is based on Pender et al. (2001j, Sserukuma et al. (2001), Bashaasha (2001) and Woelcke (2003) . Land use trends and SLM trends are adapted from Pender et al. (2001), who collected survey data in 107 communities across Uganda during 1999-2001. Figure 17: Land degradationhotspots and development domains in Uganda Development Domains in Uganda Northwestern Region 4 Populat~on Market Rainkd agricultural dsnsrty ac-5s potential low low Unimodal - very low low low Unlmodal - low low low U n i m d a l - medium low h gh Unimodal - low low low U n ~ m d a l high - low hi&I Untrnodal - medium low low Bimodal - low i'" z . 3 low low Bimodal - medium low hsh Unimcdal - high ? .&. low - , ---! high Blrnodal low rn h~gh low Untmodal - vary law low Bimodal - high .--- low - ---2 low h~gh Bimodd - medium Sou r h 9 h low Unimodal low --- low Un~rnodal med~urn - --. hl9h ---1 low high Bimodal - high : hdl high Un~rnodal low I h* low Unimcdal -- hlgh hl high Unimodal - medium f*,, heh low Bimodal - low low BimodalUnimodal medurn - - high high high &modal - low Fm h~gh low Bimodal - high kZEB high high Bimodal - medium EEm h 9 h h1gh &modal - h(gh a Hlghlsnds Lek es .=7. National Park \ I i NO Data Southwesterncattle corridor Source. Gard Ru-cker, IFPRI Source: Own presentation. 15:. Pender et al. (2001) selected three factors to determine the comparative advantage of locations across Uganda: (i) agro-ecological potential; (ii) access to markets; and (iii) population density. "Development domains" have been classified by overlaying these three din~ensions.Figure 17 illustrates to which development domains the identified land degradation hotapots belong to. Three hotspots --the Lake Victoria Crescent Region, the Southwestern Ilii,hlalids, and the Eastern Highlands - can be largely characterized by high population density, h~g'lmdrket access and high agro-ecological potential ("high-high-high"). The southwestern Cattle Corridor covers mainly the "low-high-low" and "low-low-low" development domains. Thc Northwestern Regions cover a variety of development domains, including "low-low-low", "lo .Y-low-medium",and "low-high-medium". 5.1 Lake Victoria Crescent Region 1 Development domain, trends in land use and SLM: In general, since 1990 this region, ~ 1 1 ~ 1 1 high ago-ecological potential, high population density, and high market access, has experienced a slight increase in banana and coffee production. In some areas around the Lake t r l ctona C'rescei~t,however, the coffee production has not increased but remained at constant lev.:l. Regarding the use of land, the area under cultivation has been grown, at the expense of grazing land and wetlands. Forest availability has been declining less in the stable coffee ptt ii~iL~i(>ilregions, and the quality of forest has been declining least where the coffee and bd~aria production was increasing. 1 15(1" Adoption of many soil and water conservation practices (SWC), such as tree planting, mulching, composting, soil bunds, and the use of animal manure, has been rising faster in areas of banana and coffee expansion compared to other development pathways. About ontb-third of all farmers producing coffee and banana have adopted one or several SLM ieclmicjues. 'She use of fallow-strips, however, is declining. Surprisingly, the use of all purchased inpats was average or below average, despite relatively high market access and a number of tcc'br:ic'il assistance programs in the region. However, there has been a major increase in the use r;+" f'ertilizer in this region since 1990, and the use of improved seeds llas also increased sig fizantly li ;bf' Potential SI,M-based development pathways: Promoting the "expanding banana and cdbfee production" pathway may be a potential "win-win" development strategy, benefiting the ell7 lrollment wh~lecontributing to economic growth and poverty reduction. Expansion of banana air1 coffee systems was most strongly associated with adoption of soil and water conservation pr;ztices, iinprovements in resource conditions, agricultural productivity, and human welfare. ?'haspdthway is not suited to all parts of Uganda, however, and has been developing most in the h3r ~oddlhigh dnd low rainfall zones. The sustainability of coffee production may be hrther lncr-eascdby exploring options of shade grown coffee in combination with other SLM practices :,ut il a. mulching. Such systems can become increasingly important in the context of climate cil<*ll~t Ir, Ilwe i~its comparative advantage, other potential pathways for the region include intr:asi~icationof other perennial cash crops, perishable annual cash crops (including non- traditional), and intensification of livestock production (including dairy production). All of these pathways are already pursued to some extent. However, there are many opportunities for grt ater sustainable i~ltensification.Increased horticultural production has been inore common in bimodal medium rainfall zones so far and has -not surprisingly-been associatedwith access to irrigation. If increased horticultural production is considered as an option it needs to be taken into account that fallow land, grazing land, and woodlots have been decreasing in such areas. Most sustainable land management practices did not change significantly for the "horticultural pathway" since 1990. 162. Priorities for public investments and policies: Promoting these potential sustainable development pathways through public resources in the Lake Victoria Crescent Region may include improving the access to markets. In general, road development and associated development of transportation and other services have contributed to improvements in many natural resource conditions and human welfare indicators in Uganda (except forest and wetland availability). Other constraints are the lack of adequate storage facilities and limited availability of electricity for cold storage. These investments are particularly important to ensure that the perishable products reach urban and export markets quickly. Investments in market information systems-facilitatedby advisory services - firther improvethe linkagesof farmers to markets. 163. Appropriate investments in irrigation infrastructure may be another entry point for public support in this region given the potential of horticultural products. Across Uganda, irrigation appears to reduce pressures to expand cultivated area at the expense of forests and wetlands and fallow strips, contributes to the adoption of fertilizer, and is associated with improvementin several resources and welfare indicators. 164. Of particular concern in the Lake Victoria Crescent Region are the extremely negative potassium balances, since banana is a heavy feeder of potassium Strategies for increasing productivity of banana-coffee systems may include the promotion of inorganic (potassium) fertilizerthrough advisory services. Fertilizer may be promoted as a complementary part of an overall strategy of integrated nutrient management, rather than a substitute for organic fertilizersand soil and water conservation.Increased commercializationin the region may enable farmers to purchase inputs, such as fertilizers and improved seeds. Application of inorganic and organic fertilizer becomes even more important, since it can be expected that increased commercialization leads to higher rates of nutrient exports. Investments in rural finance may firther strengthenthe ability of farmersto adopt purchased inputs. 165. The use of mulch and compost could be further increased, since perennial production systems guarantee the availability of crop waste. In the context of horticultural development pathways the promotion of safe pesticide use and integrated biological pest control is crucial. This can be achieved through regulatory frameworks and standards and support fkom advisory services. As for the other pathways below, capacity building of researchers and extension agents with regard to relevant SLM practices and technologies is needed. Earmarked h d i n g for research can be used to generate and disseminate relevant practices and technologies. Investmentsin research may include a focus on non-traditionalcash crops and livestock products with high profit potential. 5.2 Southwestern Highlands 166. Development domain, trends in land use and SLM: This region belongs to the same development domain as the Lake Victoria Crescent Region, i.e. high ago-ecological potential, high population density, and high market access. The region is dominated by cereal and banana production. Changes in crop production have not been significant. With regard to land use, a substantial increase in the cultivated area can be observed since 1990. Wetlands have also increased in the same period. Land scarcity is extreme in the densely populated region. Forest quality is being preserved in the Southwestern Highlands more than in other zones and the diversity of wild plants has increased as well. 167. Mulching, manuring, composting, trash lines and incorporation of crop residues are relatively common soil and water conservationpractices, practiced by about one-fourth of the farm households. The use of fallow strips and soil bunds has declined. In the latter case, farmers are reportedly destroying soil bunds to harvest the fertile soil that they contain. There has been little change regarding other practices. The use of purchased inputs is close to the national average, while use of pesticides has decreased since 1990. Government and non-government programs are very common in this region. Most of them are focusing on income generation, poverty eradication, and social development, with only a few programs focusing on SLM and environmental issues. Some NGOs in the Southwest are promoting integrated pest management and other low externalinput approaches. 168. Potential SLM-based development pathways: Potential pathways for the region include intensification of perennial cash crops, perishable annual cash crops (including non-traditional), and intensification of livestock production (including dairy production). Banana systems are reported to achieve higher yields than in the Lake Victoria Region. Further sustainable intensification and expandingbanana-coffeeproduction systems may be one promising pathway. Growing domestic demand for livestock products, vegetables, and fruits provides incentives to pursue the respective development pathways of horticultural and livestock production. Adoption of crossbred cattle is increasing in many zones, but not in the southwest highlands, which is an indicationthat there is scope for increasingdairy production. 169. As discussed above expansion of horticultural production has been associated with reduced fallow land, grazing land, and woodlots in Uganda. Given the fragile ecosystems of the highlands, SWC practices need to play an important role. However, it needs to be ensured that all promoted technologies and practices are financially attractive to the farmer. As an example, SWC could be stabilized by Napier grass or Calliandra which would reduce maintenance (labor) costs, increase soil nutrient inputs, and can be used as dairy meals or fodder. Mulching and incorporation of crop residues increased slightly (but not statistically significant) and may be options to improve the management of natural resources. Agro-forestry systems as intercropping and boundary planting (including fruit trees) and manure application are m h e r options in the context of "high-high-high" developmentpathways. 170. With regard to livestock production the impacts on nutrient depletionand soil erosion would need to be observed carefully. Where dairy development is occumng, there are likely greater opportunities to promote zero grazing livestock systems linked to intensive crop production and based on confined feeding and recycling of animal waste than in areas where more extensive livestock production is practiced. There may be opportunities for increased integrationof crop, livestock and ago-forestry activities. 171. Priorities for public investments and policies: Investment opportunities are similar to those of the Lake Victoria Crescent Region. A necessary requirement for the above pathways is increased market access, which is simultaneously associated with improved natural resource conditions. One of the major constraints for horticultural production in Kabale is poor transport and marketing infrastructure. Roads and storage facilities are required to take advantage of the temperate climate in the western highlands to push the production and marketing of horticultural crops. There is also the need to develop farmer cooperative production and marketing in order to overcome some of the marketing constraints. Improving the quality of livestock herds seems to be another entry point in this zone. Support can be provided to institutional innovations for market development, e.g. contract farming, outgrower schemes, and dairy cooperatives. Again, the development of market information systems and rural finance institutions may yield high social returns and facilitate a process of sustainable development. Further capacity building is needed for regulation and monitoring, which are crucial institutional elements for agricultural markets to function well. It may be necessary to develop official systems of grades and standards. 5.3 Southwest Cattle Corridor 172. Development domain, trends in land use and SLM: The Southwest Cattle Corridor can be characterized by bimodal low agricultural potential (low rainfall) and generally low population density.21Parts of the corridor have low and others relatively good market access. Some small pockets have high population density. Pastoral and some annual crop systems are dominant in this zone, while many areas experience an expansion of the latter. Overgrazing is a major problem leading to soil erosion, nutrient depletion, and invasion of weeds. Other common unsustainable management practices include deforestation, bush-burning, and reduced falIaw periods. 173. Potential SLM-based development pathways: Increased extensive livestock production is one potential development pathway for the comdor. Extensive production of hidl value livestock that is relatively easy to transport, such as cattle and small ruminants, can occur in areas away from markets that are low in potential for crop production. Dairy products may also be produced in such extensive systems in low potential areas, but high access to collection and processing facilities or to urban markets is essential. Decreasing nomadism, better market access, and population increases influence the evolution of the system. Increases in artificial insemination, introduction of zero-grazing and increasing food production are major changes in the zone. In areas with increasing population density and high market access, increased intensification of livestock production (incl. dairy production) is possible. Mixed crop-livestock production may develop as population density rises in pastoral systems, with farmers keeping animals for plowing, consumption purposes and as a form of saving. This is because the benefits of exploiting complementaritiesbetween crop and livestockproduction rise as population densit)- rises, particularly where markets are not well developed (McIntire et al., 1992). 174. Incorporation of crop residues is more common in less densely populated areas, such as most parts of the Southwest Cattle Corridor, probably because the use of annual crops and tillage is greater in these areas, which are drier and have lighter soils than other parts of Uganda. Where intensive livestock production is occurring, such as dairy development, increased use of stall feeding and recycling of animal wastes to the soil through manuring and composting is likely. Such changes can bring about improvements in soil fertility, though this is not assured given increased export of nutrients via commercialization. There can also be impacts on water quality and environmental conditions (particular where ago-chemical use is increasing).Despite The distict of Yi?!s:ji+ to r l , ~crlttlr ri,rridor is inchlded hrre the dwindling pasture and water resources in the overgrazed areas in the pastoral communities, there is potential for increasing the carrying capacity of the dry areas by promoting leguminous trees and pasture that are well adapted to the dry environment. Such trees (for example gum arabica) help to control soil erosion, fix nitrogen and enhance biodiversity, and provide a marketable product. In the less populated areas, increased use of fallow strips is another SLM option. 175. Priorities for public investments and policies: Policy and investment priorities should probably focus on road development, and addressing land tenure issues and conflicts over land and water use. Improved market access provides opportunities for stronger market orientation of both livestock and crop production. The need of pastoralists to have access to grazing land and water sources should be considered when investing in roads and promoting settlement in areas used by pastoralists. Improving collective action in managing communal grazing land may also be a priority. Participatoryresearch and extension approacheswill be particularlyimportant since the pastoral communities have long-term knowledge of the ecological and socio-economic environment that may be useful in promoting SLM practices and local institutions for enforcing SLM regulations. As an example, the bush burning problem is more likely to be resolved if customary institutions actively participate in enforcingthem. 176. Given the relatively low population density, investments in agricultural research and extension should probably focus more on less labor intensive SLM practices and technologies (e.g. fallow strips). In principle, the impacts of population growth on a particular type of land improving investmentdepend upon the way it affects land and labor costs, as well as labor and land intensity of the investment. Since poor breeds and poor management are leading to low productivity of livestock production, investments are needed in improving the quality of the livestock herd. Due to limited investment in livestock research and extension - less than 1percent of research and extension expenses according to NEMA (1999) - there has been little attempt to improve local breeds. Public investments may also aim at improving access to veterinary services, animal health products, and more developed market channels (including provisions for compliance with sanitary and food safety standards). Finally, given the low rainfall in this zone, improved and sustainable management of water resources on private and public land should be addressed by research, extension and at the policy level. In this context, investments in small-scals irrigationmay be considered as well. 5.4 Eastern Highlands 177. Development domain, trends in land use and SLM: High population density, high market access, and high (unimodal) agro-ecological potential are major characteristics of the Eastern Highlands. Production of cereals and bananas are the dominant farming systems, while keeping cattle is an important secondary activity. Proximity to markets in Kenya may be facilitating the development of dairy production. Coffee production is another dominant economic activity. Land is very scarce in the Eastern Highlands. Fallowing is practiced by less than 10percent of households, and the average fallow period for households using fallow has declined from 1.4 years in the late 1980sto 0.6 years in the late 1990s.Settlementsand woodlots are the only land uses that are increasing. Expansion of annual crop production is common in this zone leading to significant soil erosion. A wide variety of soil and water conservation are used. Use of grass strips, contour plowing, incorporationof crop residues, manuring, tree planting, and soil bunds are all relatively common. Use of animal.manure, incorporation of crop residues and planting grass strips have increased since 1990, while use of other SLM practices has not changed significantly. The Eastern Highlands have the highest proportion of households using purchased inputs, including fertilizer, herbicides, improved seed, fodder, animal vaccines and animal medicines. 178. It is important to note that while the rates of input use are among the highest in the country and SWC practices are relatively common, soil fertility and other aspects of land degradation appear to be worsening. This development is certainly a reflection of the special challenges the Eastern Highlands are facing, including extreme land scarcity, increased settlements, expansion of cultivation on steep slopes, and declining availability of energy sources. It also shows that the adoption of new SWC technologies is low and application of purchased inputs still needs to increase. 179. Potential SLM-based development pathways: Given the "high-high-high" development domain, a wide range of SLM-based development pathways are possible in the Eastern Highlands. Intensified cereal production, intensified coffee and banana production, 'and horticultural production - all in combination with intensification of livestock production - are among the options for this zone." In general, increased cereal production has been associated with reduced grazing land and increased woodlots. Changes in land management practices and use of purchased inputs have not been significant under this pathway. Sustainable intensification of cereal production needs to be based on improved integrated nutrient management practices, which combines inorganic and organic fertilizer and soil and water conservation. As an example, intercropping cereals and legumes have increased the productivity of cereal crops (Kaizzi et a/., 2007). Agro-forestry systems can also significantly enhance soil fertility and reduce soil erosion in cereal cropping systems. They can also be used to stabilize SWC practices and increase the financial attractiveness of such measures (see section on Southwestern Highlands). Shortage of energy sources and high market access further increase the incentives for such systems. An opportunity is certainly that this zone is close to the Kenyan border where both fertilizer and cereal markets are well advanced. 180. Priorities for public investments and policies: A wide range of investments are required to promote a sustainable intensification of crop (particularly cereal) production. The availability of improved seeds and other purchased inputs can be enhanced, and marketing of cereals could be expanded through infrastructure investments and policies that ease cross-border trade of agricultural inputs and products with Kenya. Another entry point constitutes investments in seed research, production and marketing. Currently, lack of rules and regulations are a major disincentive for the private sector to participate in the seed market. However, some grain trade companies, such as Harvest Farm Seed, have demonstrated the interest and the possibility of the private sector involvement. Input supply systems, such as nurseries for ago-forestry systems, could be initiated and strengthenedthrough matching grants or comparable financial instruments. Lack of adequate labor-saving cereal harvesting technologies constitutes another constraint, which needs to be addressed through the agricultural research and extension system. To effectively reduce cultivation on steep slopes in the Eastern Highlands, the enforcement of respective rules and regulations shouldbe done at the local and community level. 22This section only focuses on intensified cereal production. For the other potential pathways and required investmentssee section on Lake Victoria CrescentRegion. 5.5 Northwestern Uganda 181. Development domain, trends in land use and SLM: This zone has mainly low population density, low market access, and low to medium ago-ecological potential.23Some areas, however, have higher market access. In the West Nile System,cereal crop production (e.g. cassava, finger millet, sorghum, and groundnuts), as well as tobacco and cotton production, are the most important livelihood strategies. Increase of cotton and tobacco production systems experience an increase in the availability of wetlands and probably related to that an increase in the diversity of wild animal available. Fallow strips have been increased since 1990, but applicationof manure and mulch has been decreased significantly. 182. Potential SLM-based development pathways: Most realistic pathways include an increase in cotton and tobacco production, improving current subsistence crop and livestock production, and exit from agriculture by encouraging migration as well as alternative, non-farm uses for land (to conserve natural resources and promote off-farm employment). Expansion of traditional subsistenceproduction can improve food security,but are not likely to offer a realistic perspective for significant increases in household income. In areas with medium ago-ecological potential the production of both high and low value forest products should be explored. Since significant amounts of fuelwood are used for tobacco curing, increased tobacco production can lead to severe problems of deforestation. Tobacco production has already contributed to severe deforestationof indigenoustrees in the west Nile (Karugaba, 2001). Attempts have been made to require contract farmers to plant fast growing trees (such as eucalyptus)per unit area of tobacco planted. In general, this seems to be an appropriate approach to address the fuelwood demand for processing tobacco. However, the selection of tree species and the type of enforcementis crucial in this context. 183. Price changes and marketing issues are key issues for the development of cotton production. The ginners and their intermediaries have a strong monopoly for buying cotton despite the liberalizationprocess. At the same time, the American African Growth Opportunities Act (AGOA) and the European All But Arms (ABA) initiatives give African countries preferential import taxes. Ugandan cotton farmers indicated that their ability to cope with droughts is very limited. The application of inorganic and organic fertilizer has been very low in the cotton sector. Integratedpest management approaches should be explored for both cotton and tobacco systems. Integrated nutrient management systems are among the most feasible options for yield and soil fertility improvements. 184. Areas within Northwest Uganda with medium agro-ecological potential may be suitable for some forest products. Production of high value forest products such as timber may be economical in remote locations (if suitableroad and transport infrastructure exists), while low value products such as fielwood must be produced close to markets, unless they are used only for subsistencepurposes. A conversion of fielwood to charcoal can extent the marketable range of fuelwood products, however. An important prerequisite, in particular for commercial forest production, is clear regulation of land tenure. Another potential option to be explored is the production of biofiels in Northern Uganda. However, it is crucial to have adequate policy 23The District of Hoima is includedhere. frameworks, strategies, and standards in place to ensure positive effects on environmental sustainability and poverty reduction. 185. Priorities for public investments and policies: Priorities for public investment may include community-driven development initiatives to manage natural resources and generate income. The Community-driven Develapment (CDD) approach allows local institutions and organizations to develop community-led initiatives, collectively manage natural resources, and establish participatory approaches to provide services. Woodlot planting in sparsely populated areas can reduce the use of dung and crop residues for he1 and offer viable income opportunities. Hedgerow intercropping and relay cropping are productive agro-forestry systems with considerable potential for scaling up in low access areas. The conservation of biodiversity, an important asset in low potential areas, is particularly challenging yet potentially rewarding. Payments for environmental service may be an innovative approach to address this issue. Investments in small-scale irrigation may help farmers to grow some additional vegetables in the dry season. Often irrigation can be incorporated effectively in watershed management approaches in which collective action harmonizes the use of soil, water and vegetation to increase productivity and conserve natural resources. With regard to tobacco and cotton production similar investments in research, extension and infkastructure are needed as outlined above for other cash crops. Human capital development (education and training) is needed to enable migration in the medium to long run. 186. Priorities for policy should focus on institutions for collective action, sustained by a policy environment that enables group organization to function. Further, policies should :nsure equitable access to assets and support institutional arrangements to manage common pool resources when private ownership is not desirable. Regulations should provide incentives for sustainable natural resource management and provision of environmental services. Table 7: SLM-based development pathwaysfor land degradation hotspots high market access, high perennial cash crops (including associated developmentof 1management (INM) agro-ecological potential expansionof coffee-banana transportationinhastructure (bimodalhigh) system) (also to promote private 1Increase us of mulch and investmentin marketing, compost (perennials) Sustainableintensificationof storage, processing) Land use trends: high value cash crops (including Increase of inorganic Area under cultivation fiuits, vegetables,vanilla) Investmentsto increase fertilizer(in particular increasing; grazing land access to electricitylpower Ipotassium) declining; wetlands Sustainableintensificationof decreasing;forest livestockproduction(including Rehabilitationand 'Integrated pest management availabilitydeclining(but dairy) maintenance of irrigation for cash crops less than in other regions) systems to improve water allocationand water use SLM trends: efficiency Tree planting;mulching; animal manure increasing Investmentsin rural finance (used by 113of farm households);fallowstrips Market informationsystems decreasing;fertilizer (link producer to markets) increasing,but below national average Investment in research (earmarkedfunding): generationand dissemination of INMtechnologies,non- traditionalcash crops, IPM Capacitybuilding of researchersand extension agentsto develop and disseminate SLM technologiesand practices iouthwesternHighlands Support to institutional Agoforestry systems high market access, high perennialcash crops (including innovationsfor market (intercropping,boundary ago-ecologicalpotential expansionof coffee-banana development(e.g. contract planting, fruit trees) [bimodalhigh) system) fanning, outgrowerschemes; group marketing, producer- Combining SWC structures Land use trends: Sustainableintensificationof trader contracts) with shrubsltrees(e.g. napier Substantialincrease in perishable annual cash crops grass, Calliandra) cultivated area; increaseof (horticultureand non-traditional)Investmentsin transport and wetlands;preserved forest marketing idi-astructure Zero grazing (incl recycling quality; diversity of plants Sustainableintensificationof (roads and storage facilities) of animal waste) increased livestockproduction (including Integratednutrient dairy) Investmentsto increase SLMtrends: access to elect~icitylpower management (includingsoil Mulching, manuring, and water conservation, composting, trash lines, Market informationsystems mulching and incorporation incorporationof crop (linkproducer to markets) of crop residues) residues relatively common (114 of farm Supportdecentralized Integratedpest management households). enforcementof NRM Use of fallow stripsand regulations Increase of biorganic soilbunds declined fertilizer Investmentsinto improved livestockbreeds Capacitybuilding for regulationand monitoring (systemsof grade and standards) Investment in research (earmarkedfunding): generationand dissemination of agroforestrysystems,INM technologies, non-traditional Capacitybuilding of researchersand extension agents with regard to relevant SLM technologiesand ~ractices Southwestern Cattle Low population density, Increased extensiveand Improve access to veterinary Leguminous trees and ;,llr:rridor(including, high market access, low intensificationof livestock services,animal health pastures adapted to dry agro-ecologicalpotential production products, more developed environments (bimodal low) market channels with Sustainableintensificationof enforceableregulatory Increase use of fallow strips Low populzdoc density, livestock production systems to ensure compliance low market access, low with sanitary and food safety Zero-grazing (if culturally agro-ecologicalpotential Increased (sustainably standards) acceptable) (bimodal low) intensified)crop-livestock systems Improve access to collection Manure applicationand High populatim density, and processing facilities for composting (if animalpower high market access, low Exit ag-iculture(medium to dairy production; storageand can be used for agro-ecologica; potential long-tern) by encouraging transportationof perishable transportation) (bimodal low) only - migrationand other non-farm goods smallpart use of land (to conserve natural Incorporationof crop residues resources and promote off-fm Investmentsinto participatory employment) research and extension Crop-livestock-agroforestry Land use trends: approaches(based on long- systems( I M ) Deforestation,reduced term lcnowledge of pastoral fallow periods, communities) overgrazing,bush burning Enforcement of SLM SLM trends: regulationsthrough local Generally very low institutions (e.g. bush adoption, some burning) incorporationof crop -- residues Investmentsto inqrove - quality of carrying capacity of rangelands (e.g. by promoting well-adapted leguminous trees, shrubs and pasture) Investments in small-scale irrigation Investments in roads and market access (however, access of pastoralists to land and water sourcesneed to be ensured) Improving collective action in managing communal grazing land Human capital development (education and training) to foster migration Capacity building of researchers and extension agents with regard to relevant SLM technologies and practices Eastern Highlands High population density, Sustainableintensification of Market infrastructure INM (including intercropping high market access, high cereal production developmentfor cereal and cereals and legumes) agro-ecologicalpotential legume production (unimodalhigh rainfall) Sustainableintensification of Increased use of inorganic horticulturalproducts (and other Policies to improve cross- - fertilizer (close to Kenyan perishable annual crops) border trade with Kenya market) Land use trends: (input and output markets) Settlementsand woodlots Susminable intensification of Increase of woodlots and are the only land uses that coffee and banana production Investments in seed research, agroforestry systems are increasing production and marketing (including h i t trees) Sustainableintensification of (improve incentives for SLM trends: livestock production private sector through plant Combining SWC structures Declining fallow; wide varietyprotection law) with shrubsltrees (e.g. napier range of soil and water grass, Calliandra) conservationpractices Research and investments in used (grass strips, contour labor-saving cereal harvestiilg Improved cereal seeds plowing, incorporation of technologies crop residues, manuring, Zero-grazing and promotion tree planting, and soil Strengthening availability of of recycling of animal waste bunds); highest proportion agroforestryplanting of households using materials and other inputs Manure application (transport purchased inputs; (e.g. nurseries for agro- must be availableplus however, adoption of new forestry systems) through affordable labor) technologiesand practices matching grants and slow comparable financial Mulching for perennial instruments systems Capacity building of researchers and extension agents with regard to relevant SLM technologies and practices For additional investments see other high-high-high development domains - -- NorthwesternRegions Low populabon dens~ty, Sustainable intensificationof Support to CDD activities for Woodlots (fast growing (includingHoima) low market access, low cotton and tobacco production collective managementof indigenous trees) and 9-0-ecolog~calpotential . natural resources and .~ . .. -~ agoforestry--s . e m s to :unimodallow) Improvingcurrent subsistence provision of services crop and livestock production curing To some extent: Support implementationof Low population density, Production of high and low payment for environmental Hedgerow intercroppingand low market access, value forestproducts servicesfor watershed relay cropping medium agro-ecological management and biodiversity potential (unimodal) Exit agriculture(medium to protection tntegrated nutrient and long-term)by encouraging management Low population density, migration and other non-farm Investmentsin small-scale high market access, used of land (to conserve natural irrigation tncrease use of inorganic medium agro-ecological resources and promote off-farm fertilizer, particularly for cash potential (unimodal) employment) With regard to cotton and crops tobacco similar investments are needed as listed for other tntegrated pest management Land use trends: cash crops above for cash crops Deforestation, increase in wetlands and diversityof Human capital development Conservationof biodiversity wild animals(other land (trainingand education)to (throughPES) use formsdecrease) enable migration SLM trends: Policies should ensure Fallow strips increased, equitableaccess to assets and but applicationof manure supportinstitutional and mulch has been arrangementsto manage reduced common pool resources (includingland tenure issues) Investmentsin research and extensionto generate and disseminate relevant technologiesand practices Capacitybuilding of researchersand extension 6. POLICY CONCLUSIONS AND RECOMMENDATIONS 187. This section makes some key policy recommendations on how to increase allocative efficiency and effectiveness of public SLM expenditure in the short to medium term and how to identify entry points for public support. In general, the strategy could be to undertake some immediate measures to increase the efficiency of existing resources. In addition, some practical recommendations and guidelines are identified which are relevant for medium-term decisions.Of outstanding importance,however, is rigorous monitoring and impact evaluation to strengthenthe case for SLM and to inform the allocation of (additional)resources in the future. M&E as the most important step towards increased efficienw and effectiveness: 188. StrengtheningM&E is crucial to make a stronger case for SLM, inform the allocation of public resources in the future, and is a prerequisite for scaling-up SLM activities. In negotiations regarding strategic priority setting and budget allocation, it is important to show: (i) which interventions work on the ground to solve socio-economic and environmental problems; (ii) how much they cost; and (iii) what rates of return they have in terms of contributing not only to environmental sustainability,but also to productivity growth. So far, the limited availability of such types of data has been a challenge for ensuring an adequate budget allocation for SLM in the budget process. 189. Immediate action could be taken to assess the impacts of past and on-going SLM projects based on availablesurvey data. This would be the fastest way to provide the urgently needed evidence of SLM outcomes and impacts. Opportunities should be explored to base the evaluation on various plot, household and community-levelsurveys conducted by UBOS, IFPRI, and others in the recent past. These institutions did a series of surveys since 2000 and they may constitute a good basis for evaluation of SLM impacts. 190. The development of a country-wide SLM M&E System could start immediately. Initial steps may include: (i) a readiness assessment to clarify roles and responsibilities, analyze the quality of existing structures and identify capacity building needs; and (ii) finding an agreement on the outcomes to monitor and evaluate and their performance indicators. Designing and sustaining a sound results-based M&E system (including impact evaluation) is a complex task, particularly with regard to SLM. Identification of an appropriate institutional set-up, data collection and analysisrequire strong analytical skills. Hence, the development of a M&E system for SLlM should not be outsourced completely to external experts, but rather conducted in a collaborative and participatory manner in order to build sustainablein-country capacity. 191. The following key aspects should be considered while designing SLM M&E systems and conducting rigorous impact evaluation: (i) financial attractiveness of SLM technologies and practices for land users is a necessary condition for scaling-up and should be captured as a key element; (ii) explicit consideration of externalities may help strengthen the case for public SLM support and design payment for environmental services systems; (iii) spatial scaling-up of SLM in land degradation hotspots on a pilot basis should be accompanied by M&E implementation from the start; (iv) the M&E system shouldbe designed in a way that can help to identi@ the most efficient and effective types of public SLM interventions. Short-term measures to increase efficiencyand effectivenessofpublic SLM resouycs 192. Closing the gap between budgeted and actual expenditure can significantly increase public SLM spending in the short run. As mentioned in the findings ahrwe, actual disbursement was only 42 percent of planned expenditure. Reaching the full ceiling of allocated SLM funds would have increased actual expenditure by UGX 19 billion per year between LOO1 and 2005 already. To address this gap - an issue which is common across many sectors in Uganda - a strategic review of procurement and disbursement processes and procedures would be required. Clarification on the priorities for public support may further increase disbursement (see section 4.2 for details). 193. Spatial targeting of SLM expenditure can be further improved by focusing on land degradation hotspots as the main allocation criteria. While public resources have been targeted relatively well spatially, there is still room to increase their effectiveness and efficiency. The Eastern Highlands, the area around Lake Albert, and small pockets in the Central. Region have received low levels of public SLM resources in spite of severe land degradation. A couple of districts in Northern and Central Uganda with relatively low degree of soil nutrient depletion and soil erosion received significant public resources to address SLM issues. If allocationswould be revised acc;ordingly, the chances of successfully preventing and reducing lavd drgradation would increase (see section 4.3). 194. Various measures can be implemented in the short-term to reduce the mismatch between the goals as expressed in sectoral investment plans and actual expenditwre. Actual expenditure reveals that the concept of placing SLM in the productivity and economic growth agenda as envisaged by relevant strategies and investment plans has not yet been ?r;lnslatedinto operntionalreality. The following activities could address this issue: De\sefo~inga shared understanding of SLM at the policy and operatio~uiI~vel.This SLM concept should be as operationally relevant as possible and take ixrto account country-specific conditions. Further, the concept should clearly acknowledge the role of SLM with regard to productivity and economic growth. The concept outlirlctl in chapter 4 may lnfonn this process. a Integratzng SI,M issues in budget planning documents. These would include the upcoming National Development Plan, Budget Framework Papers, 1MAAIF's Dcvelopmellt Strategy and Investment Plan (DSIP), and all relevant Sectora1 Ix~vestment Plans. NEMA has developed Guidelines for "Mainstreaming Environn~entalIssues into Budget Framework Papers" (NEMA, 2007). This mainstreaming provision is potentially ;I very powerful tool to address cross-cutting SLM issues in the budget process. However, since the mechanism is new, it is important to assess its effectiveness (see chaptcr 3.4). * St,rengthming the role of commzinities and SLM-related Civil Socicrljl Or.~rinizations c ' ( : C 0 . ~ ) in the budget process, as well as in project preparation and in~plenlentation. Fanners and communities have repeatedly raised their concerns with regard to land degradation As the experience world-wide has shown, civil society participation can play a key role in bringing sustainability issues on to the political agenda. Yet, civil society organizations in Uganda are concerned about the problem of "token participation" with legard to their involvement. To address this problem, it may be useful to cl,42ilkthe role of' civil society participation in different processes, and to improve ;ru l)~lntzbility rileclaaiisnl~for participation at the same time (see chapter 3.4). Medium-term measures to increase efficiencyand effectivenessofpublic SLM resources: 195. Systematic integration into major land productivity programs may constitute a low cost option to srisle-rep 81 s~abially.in particular in the land degradation hotspots. Appropriate scaling xp :2:c,:! sit;Altk , f s i i \ ctllclss have not been identified or utilized yet. Such vehicles may include all major prolluctivity progr'ms, such as agricultural advisory services (NAADS), agricultural research (NARO), agricultural land management, and large-scale watershed approaches. Regarding the NAADS program, for example, the integration of SLM in agricultural advisory services should becolne an effective and efficient way of helping f m e r s to get acquainted with the results OF Sl,V research (generated by NARO and others), understand the reasons aqd outco~alr.;2 i * ;;].id ~rc,graitatiol?,ancl adopt and maintain profitable SLM practices. Currently niost agjicuiturai projects tha~szck to disseniinate farm technologies have not explicitly included SLM elelnenrs, rest~it~r~gin limited success to prevent and mitigate land degradation and thus bring z long-tern1 sustailiability to the current farm techniques. However, such programs would constiture an appropriate vehicle for scaling-up sound management of natural resources (see section 3.4.3 and chapter 4). 196. Public expenditur.: should he used mainly to encourage more private SLM investments, 'i't1e UT i.~iitii!it'. .:ijf GY ;.e,:i:ei;, to make in~restmentsin profit-generating private goods, such as in inputs ileedeii LO adopt and maintain financially attractive SLM practices. The public sector's role is to crcate an environment that favors these private investments. Examples for public sector roles may include: (i) reducing high up-front costs for SLM practices and technologies by generating and disseminating low-cost solutiolis via research and advisory services (NAADS and NA KCE): (ii) promoting community-driven development approaches based on matching gmnb, ic.3. 'I-,: .,el:.)j.~ri.!t>rlt .' ~f i:oll-in~tl~~ityni~rseriesfor agroforestry systems); (iii) providing the r.egulatory a11rj i ~ i s:r:ilioa.ialkamework for payment for environmental services ! systems; and (iv) improv:i:g access to in-kind and in-cash credit. On-going and upcoming investment operations supported by the World Bank and other development partners, such as NAADS and the Environment Sector-wide Approach, con-stituteopportunities to consider some of the entry points identified :I; this stuiiy and maivstream SLM in the land productivity agenda. 197. The idrnt(flrntlf:l: :.''+r.:~ ;;r~alji:;tc sffifncient a ~ effective SLM public expenditure d csn . 1 3 ~f C;kceta11euI .xo1l1irrcridati017~on priorities for public investments and j policies for each \';;;ii il::c;,aciat~..,,~ i i r t t r . 1r?Ls are summarized in Table 7 (chapter 5), including r~rhli - 03 ' 198. While land use sectors contribute to land degradation, the auswer is not to slow down their growth, but to seek more sustainable prodlittion systenrs tlarorlgh public and private investments and to enhance the sectors' provision of env!rcrarn~entalcervices. h Uganda, land degradation is most severe in areas with high agrchecologi6*alpoteiltial r;!ld often caused by high land use intensity. Given the crucial role of the latad trse sectorc h r ecoltomic growth and poverty reduction and given the limited scope for bringing addiriorlal lanil illlo production, the response should not be to slow down the productivity growth ir! these sectors but to base the urgently needed smallholder productivity revolution in I Jgarxda ilri 3 ~echralogychange that systematically integrates SLM aspects. While the major vole cf :he public sectcar is to crowd in private investments by providing public goods, iinancial attractiveness of SI,M practices and technologies is a key necessary condition for adoption at tgnn and community level. Integrated nutrient management is among the most promisiing kcbr:lca! options. which are financially attractive to the farmers and reverse land degradation 199. Increased public attention on climate change coeQtibe used to further strengthen the focus on SLM and tap into additional financial resources more strategically. It is widely acknowledged that SLM has an important role tra ?Iaj~with regal63 to both climate change adaptation and mitigation. Hence, SLM offers tlae rathcr unique oppcirtunities to use synergies between adaptation and mitigation and access additional financing. Currently funding for adaptation is available under three different IJNFCC(_' Funds and GEf: M i l e the resources available are limited currently, it can be expected tha? funds will increase in the future. In addition, efforts can be made to access carbon finance llnder the C '1e:in I3evelopment Mechanism of the Kyoto Protocol and the voluntary market Annex 1: Sectoral composition of the "SLM sector" and "SLM related sector" expenditure, total for 2001102-2005106 I I Soil conservation and agro SIDA Gd89 MAAIF Countqwde 0 0 forestry N-W smallholder Various ADF 0088 L-0483 -IF 55,962 5,882 3,167 agricultural development districts - Lira, Kasese GoU 0089 G-1643 Support for irrigation -IF 0 0 2,224 and Tororo GoU/KDRC/ AG Crop product' ?nsystem CrATm G-0464 h4AAIF C o u n m d e 0 0 657 research A0 21(B) ARTP 11- AG Farmpower and s&ll Various G-0464 NARo 714 0 0 IDA 75(B) irrigation systemresearch districts AG Outreach and partnership IDmP G-0464 NARO Countrywide 6,657 0 C 28(D) ..tiatives -- Conservationof agricultural FA0 G-1924 MAAIF Countqwde 0 0 0 pH, -- Piloting conservation FA0 h4AAIF Countqwde 809 809 0 -. Kawanda AG Soil fertility enhancement CIAT G-1641 NARO agricultural 0 8 1(A) research research -- NARO Counlr--de 14,711 1 7,465 9,119 1 I NARO Countrywide 591 N^IRO Coontqwde 405 176 -- Small scale irrigation 174 Forest management and MLWE EU NR 100 G-1511 Countrywide 15,923 15,923 0 0 15,923 15,923 -4 conservation Farm income enhancement MLWE ADF 0947 G-1866 Countrywide 0 0 0 0 0 0 and forest conservation Farm income enhancement NDF 0947 L-0538 Countrywide 0 0 0 0 0 0 and forest conservation Farm lncome enbancement NARO ADF 0947 L-0526 Counmde 0 0 0 0 0 and forest conservation O I ---I Various I GoUEU G-0862 Tree seed center MLWE 2,627 0 72 55 2,699 5-4 03 districts -- 1 1 NR Forestproducbon system Norway G-l NFA Countrywide 433 50 415 180 848 230 1 - 13(a) research -- --1 EIJ 0161 G-1735 National Forest Authority NFA Countrywide 0 0 0 0 0 0-I 0 - Norway 0161 G-1737 Nations! Forest Authority NFA Countrywide 1,943 1,943 0 1,943 1,9j!-1 ] Norway Supportto National Forest / 0161 G-1752 MLWE Countrywide 292 292 0 0 292 212 b- AUthori - - -4 NR Nyabeya and IYorway G-0851 Combined forestry hining MLWE 0 0 0 0 0 i I) --- 46(D) Maslndl - - -i NR Developmentof natural Eastern I 0 13,950 10,Ois I I Narway G-095 MLWE 13,950 10,013 1 0 - -+ 46(A) resourccs/forestry Uganda - I I 1 Supportto Nabonal Forest D m 0161 G-l734 NFA Countrywide 0 0 ,--- Authority I NR Central forest G-0373 Forestry rehabll~tation 0 0 -- h >way G-0193 Nabonal 2,612 953 205 2,817 --- -..A - Nyabushozi AG 62 Pasture farming in Ankole GoU/GTZ 0-0912 Mbarara county, 0 273 41 (A&B) ranch Mbarara GTZ G-1405 Integratedpasture farming MAAIF Southwest 273 273 0 I :@- ( - I I AG Animal production system Various 1 1 ( 6,546 0 151 74(B) research districts . NARO I1 -I I I I I I I:INTBRATEDWATERSHEDMANAGEMENT 1 63,605 (9,019 36,400 various 0169 G-I328 Water for production MLWE 3,132 0 26,116 -- districts Operationalwater resomce MLWE G-l 85 management-Nile Basin Countrywide 7,863 384 1,460 Supportto water resource MLWE 0163 G-9922 monitoring department Countrywide 10,936 9,776 4,935 ~hase111 - Livelihoodlake dependent MLWE DFID G-1301 Countrywide 0 0 0 communities Mt. Elgon conservationand MLWE Mbale and NR 46Q G-095 0 1,207 0 development Kapchorwa - NR 44 Lake Victoria environment MLWE Lake Victoria IDNGEF L-0436 ,609 547 (ABB) management basin Wetland sector strategic MLWE BELGIUM 0146 G-1733 Countrywide 4,410 0 706 lannin su EGYPT/Go MLWE u ;\G3: G-1211 Water hyacinthprogram Lakes 6,519 25 35 A&B 7 NETHERL National wetland program MLwE 1 1 G-1021 ,, 1 1Countrywide 7,584 756 1 2,481 ANDS A&B 07) Reducing biodiversity loss IDNGEF 2 ) G-0970 at cross border sitesin East NEMA districts 1,569 1,846 120 Aiiica Conservationof critical USAID G-l NEMA Southwest 7,342 7,342 0 ecosystems SIDA G-0521 Lake Mburo OPM Nyabushozi 0 0 0 county DFID G-1290 Lake resource management MLWE Countrywide 12,643 9,395 0 0 12,643 9,395 Productiveresource MLWEl Southwestand USAID investmentfor managing the 0 13,107 0 0 0 13,107 MTTI west environment CATEGORY5: ENERGY-RELATEDACTIVITIES 7,172 8,476 0 0 7,172 8,476 Various SIDA 0092 G-1548 Rural electrificationI I 7,172 7,182 0 0 7,172 7,182 wF districts AG Lake Victoria fisheries Lake Victoria EU G-0304 NARo 0 0 0 0 0 0 11(G) research basin AG Lake Victoria fisheries Lake Victoria EU G-0888 NARo 7,944 4,925 0 0 7,944 4,925 11(G) research 2 basin CATEGORY7: PROTECTEDAREA MANAGEMENT 89,556 40,375 10,667 8,204 100,223 48,579 * KibaleISemulikiforest Bundibugyo, G-1144 conservationand MLWE Kabarole and 0 0 0 0 0 0 ANDS 06(A) developmentII Kasese * KibaleISemulikiforest Bundibugyo, G-1145 conservation and MLWE Kabarole and 3,354 1,729 0 0 3,354 1,729 ANDS 06(A) development111 Kasese Western rift UNDP G-1847 Albertine Rift Biodiversity NEMA 544 544 0 0 544 544 * NR ) 52(A valle Conservationbiodiversity Western rift UNDP G-1346 1,846 1,846 0 0 1,846 1,846 52(A) rift valley NR Conservationbiodiversity Western rift GEF G-l 654 NEMA 0 0 0 0 0 0 52(A) Albert valley valley GTZ G-1775 Advisor to Uganda Wildlife UWA Murchison 1,113 1,113 0 0 1,113 1,113 N N Q m m * rt \D I- C r ) - c ' : b m W t4 O N - O W N kc'-? O Q I C IA 0 \C - w e Annex 2: Methods for predicting soil erosion and nutrient depletion 1. Econometric methods were used to estimate the spatial distribution of land degradation in Uganda. Both proximate (biophysical) and underlyin (socio-economic) causes of land degradation are used in the prediction model! The proximate (biophysical) factors include topographical characteristics such as slope, elevation, rainfall, land cover, climate and soil erodibility (Voortman et al., 2000). Climate affects land degradation for example via temperature and rain storms, both of which increase soil erosion in the tropics (Zake et al., 1997; Magunda and Tenywa, 1999; Voortman et al., 2000). Soil physical characteristics affect the erosion susceptibility. Soil erodibility depends on topsoil texture, shear strength, aggregate stability and organic matter content. Unsustainable land management practices change the proximate characteristics desciibed above and lead to soil erosion, which in turn contribute to soil nutrient depletion. 2. The underlying (socio-economic)causes of land degradation are numerous and their relationship with land degradation is complex and context specific in many cases. Underlying causes include population pressure, poverty, land tenure, agricultural commercialization, level of input prices, and access to rural services (which include advisory services, financial services, marketing, roads, transportation, and other services that affect rural livelihoods and agricultural production). It is also hypothesized that the interaction of some key variables may affect land degradation in a way that is different from the individual factors. Soil erosion and nutrient balances are used as indicators of land degradation. Both soil erosion and nutrient depletion are affected by the same factors, since soil erosion is one of the major channels of soil nutrient outflow (Smaling st al., 1993). Procedure ofprediction of land degradation 3. Household and plot level surveys conducted by IFPRI in 199912000 and 2001 are used for the analysis. These data are complemented by another survey canied out in 2003 by IFPRT and the Uganda Bureau of Statistics (UBOS). The three surveys collected plot and household level data for estimating soil nutrient flow and soil erosion. Data were collected from 1,426households in 270 local community councils level 1 (LC1) based or1 stratified random sampling. The stratification was based on indicators of agriculimal potential, market access, population density and altitude (Nkonya et al., 2004; Pend~r.,,a ul., 2004; Nkonya et al., 2005b). All plots operated by the selected households wsre surveyed and the total number of plots surveyed was 5,391.~~ 4. Plot level and community or higher level data are used to estimate the coefficients of the variables selected as the proximate and underlying causes of land degradation at an aggregated level. These coefficients are then used to predict land degradation in each of 24For detailed discussion of the proximate and underlying causes of land degradation in Uganda, see Nkonya et al. (2004). 25For details of the survey methods, see Nkonya et al. (2004), Nkonya et al. (2005b), and Pender et al. (2004). the soil mapping units (SMU) that matched the surveyed plots.26For the SMUs that were not covered by our surveys, land degradation is predicted using spatially disaggregated secondary data that matched the SMU. The secondary data was obtained fiom the Uganda 2002 Census (including data on poverty incidence, population density; and urban and rural location). Data for land cover, travel time to main town, mean annual rainfall, slope and elevation was obtained fiom regional and global GIs data. 5 . As observed above, the estimates are extrapolated to locations, which were not covered by the surveys (e.g. the Kalangala, the northeastern districts such as Moroto and Nakapiripit). Extrapolation of results to these areas obviously produces less reliable predictions. Extrapolation of land degradation results to two districts (Kotido and Moroto -andNakapiripit for somedata)was notyossibledueto limitedareacoveredby SMUs that matched with those covered in the study by Nkonya et al. (2005b). 6. Figure 2 shows that some districts have various levels of land degradation. To derive the land degradation at District level, the weighted SMU average is used. This process resulted in one map indicating the degree of land degradation at district level, allowing the analysis of spatial targeting of SLM exper~ditrlre(Figure 12, Figure 13, Figure 14,and 26SMUs are classified according to their soil textural characteristics (coarse, medium and fine) and slope of the dominant soil unit (gently undulating - with slope less than 8percent; rolling to hilly - slope 8 percent -30percent and steeplydissectedto mountainous-with over 30percent slope).Excludingwaterbodies, 25 SMUs were identified in Uganda. Aggregation of land degradation within each SMU produced more accurate results than aggregation using ago-ecological zones, since the SMU capture explicitly two additional proximate causes of land degradation (soil texture and slope). SMU also represented a much finer resolution of the estimation than the agro-ecological zones, and it was assumed that the coefficient and variance does not vary within one SMU. Figure 16). 7. Obviously, in areas covered by the surveys, the predictions are subject to sampling and other estimation errors and they are constrained by the fact that the surveys covered only a single period. Long-term soil nutrient balances and erosion can be determined more accuratelyby collecting data over a longer period of time. Hence, the results should be interpreted and used, taking these constraints into account. However, as noted above, the estimates compare well with other studies and capture the general picture of land degradation in Uganda. Model specification for predicting land degradation 8. Following the discussion above, land degradation was predicted using four regression equations: Pbal = f(X,Z, V, e) ....................................................... (2) Kbal = f(X, Z, V, e) .......................................................(3) Erosion = f(X,Z, V, e) ...................................................(4) where Nbal = nitrogen balancelyear; Pbal = Phosphorus balancelyear; Kbal = Potassium balancelyear; Erosion = soil erosionhdyear; X = a vector of proximate causes of land degradation, namely elevation, slope, rainfall, and land cover; Z = a vector of underlying causes of land degradation, namely: travel time to main town, location of farm (urban = 1,rural = 0), population density, and poverty incidence; V = a vector of interaction terms of selected variables, namely population density x access to market, rainfall x population density, and rainfall x access to market; and e is an error term, which is assumed to be independent and identically distributed normally with a mean of zero and standard deviation of a, i.e., iid-N(0, a). 9. The selection of explanatory variables for soil erosion is based on the Revised Universal Soil Loss Equation (RUSLE).~'Soil nutrient balances are computed using the nutrient flow analysis, where the inflows and outflows are calculated using detailed data including laboratory analysis of soil samples collected from farm plots.28 Since soil nutrient balances are affected significantly by soil erosion, we use the same variables as in the soil erosion model to predict soil nutrient balances.29Secondary data for most of the inflows and outflows are not available. Hence, the soil nutrient estimates are likely to be less accuratethan soil erosion estimates. 27Revised Universal Soil Loss Equation (RUSLE): A = RKLSCP, where A = amount of soil lost per unit area; R = rainfall - erosivity factor; K = soil erodibility; L = slope length; S = slope steepness; C = land cover - management; and P = support practices, namely soil and water conservation structures (Renard et al.,(1991). 28For details of estimating soil nutrientbalances, see Smalinget al. (1993) and Defoer et al. (2000). 29Soil erosion is one of the sources of nutrient outflows, hence must be estimatedbefore one can estimate nutrientbalances. 10. It is tested for multicollinearity by using the variance inflation factor (VIF). Multicollinearity is not a major problem since the maximum VIF is below 10 (Mukherjee et al., 1998). However, some determinants of soil nutrients are extremely skewed. Since nutrient balances for many observations are zero, the use of log transformation, suggested in Mukherjee et al. (1998), is not possible. Thus, we dropped the outliers to allow the appropriate distribution. In the case of soil erosion, where all observations were positive, we log-transformed all continuous variables and did not drop outliers. 11. Next, we look at the functional relationship between the independent and the dependent variables. In the model specification, we use both linear and quadratic specifications of elevation, slope, rainfall, travel time to main town, and population density. All four models had higher predictive power when the quadratic terms were included in the model. A joint Wald Test is performed to determine the relevancy of the quadratic and interaction terms. Since the null hypotheses, that they are equal to zero, is rejected, they are included in the models. Based on the methodology described above, the spatial distribution of the severity of land degradation in Uganda has been obtained. The overall result is illustrated in Figure 2 and leads to the identification of land degradation hotspots in Uganda. Annex 3: Budget Processes at the National and Local Levels in Uganda 1. Against the background of the above-presented SLM strategies, policies, and major government institutions, this chapter describes and analyzes the public finance management framework related to SLM expenditure at national and local levels. The analysis of budget preparation and execution at national level are followed by the analysis of the linkages to the budget processes at local levels. Since SLM cuts across different sectors, this chapter also reviews the coordination mechanisms that exist between the different organizations involved in budget processes related to SLM. After that the chapter addresses the opportunities and challenges to the coordination mechanism for SLM issues and expenditure, looking at institutional bodies, ad hoc mechanisms, and mainstreaming provisions. The chapter ends with policy recommendations of how to improve the budget processes related to SLM. 1. Budgetprocesses at the national level 2. The budget process is governed by the Budget Act of 2001. This Act creates room for the involvement of the Parliament in the budget preparation process and specifies the timeline of the process (Table 8). Since all Ministries and Agencies are subject to the same budget process, the process is similar for all public sector organizations involved in SLM. Self-accounting agencies, such as UWA and NFA prepare and execute their own budget. Table 8: Involvement of the parliament in the budget process and time line Department, Institution,Organization or Commission shall February submit to the President preliminary estimates of revenue and expenditure Three-year macro economicplan and programs for the Not later than economic and social development in preparation for final April 1st Isubmission to be presented to ~arli&eni;indicative I - 1 preliminary revenue and expenditure for the next financial year to be presented to Parliament Sessional Committeesof Parliament discuss and review Not later than 251hof indicativebudget allocationsand submit report to Budget April Committee of Parliament I Budget Committeescrutinizesthe estimates and the reports of Not later than 151hof the Sessional Committees and submits its recommendationsto May the Speakerwho shall send the recommendationsto the President Estimates of revenue and expenditureof Government for the Not later than 15'~of next FinancialYear to be presented to Parliament June Policy Statementof each Ministry submitted to Parliament By 30" of June (should reflect data on value for money and the extent of achievement of the objectivetargets on money received and Ispent. Source: Budget Act (2001). 1.1 Budgetpreparation Planning basis 3. The planning basis for the budget is constituted by the Poverty Eradication Action Plan (PEAP), and the Sector Investment Plans, in case the sectors have developed these plans. The Medium Term Expenditure Framework (MTEF) is derived from the PEAP. As mentioned above, the Environment and Natural Resources Sector has developed a final draft for a Sector Investment Plan in July 2007. Also discussed above were the Development Strategy and Sector Investment Plan of the agricultural sector (published in May 2006) and the National Forest Plan (2002). 4. Uganda plans to move from the PEAP, which has a 3-year horizon, to a National Development Plan (NDP) with a 5-year horizon. The NDP is expected to be completed by mid-2008. It is expected to be subject to a mid-tern review, but not to annual updates. Steps in the budgetpreparation process 5. The different steps of the budget process can be summarizedas follows: Budget Call Circular: The annual budget planning process starts with the Budget Call Circular (BCC), which is usually issued in October. The BCC specifies indicative ceilings for each sector, and for Ministries and Agencies. To some extent, ceilings are also fixed for units within Ministries. For example, NAADS has its own indicative ceiling. The BCC also specifies the guidelines that the Sector Working Groups should use in preparing their Budget Framework Papers. Budget Conference: In November or December, the MFPED organizes a Budget Conference, which lasts one or two days. At this Conference, MFPED explains which resources will be available, presents the review of past performance, and explains the indicative ceilings for the different sectors. The conference is attended by members of the public administration, non-governmental organizations, and local governments. More than 1,000people usually participate in this conference. As further detailed below, budget conferences are also held at the regional level to inform local governments and stakeholders about the conditions for the budgeting of the next Fiscal Year. Development of Budget Framework Papers: Based on the BCC, the Sector Working Groups develop Budget Framework Papers. The Framework Papers are based on budget planning processes within the respective Ministries and the self- accounting Agencies. Within Ministries and Agencies, Departments prepare budgets based on their activities, which are then consolidated within the respective organization. The Sector Working Groups are expected to ensure coordination between different public sector organizations in the same sector and to leverage the participation of the private sector and civil society. The Budget Framework Papers are then discussed with the Ministry of Finance, first at the technical level, and subsequently at the Ministerial level. National Budget Framework Paper / Cabinet Vote Paper: MFPED prepares the National Budget Framework Paper and the Cabinet Vote Paper. The Cabinet provides feed-back to MFPED, which makes changes accordingly. Deliberations in Parliamentary Committees:The approved version of the Budget Framework Paper is submitted to Parliament, usually in May. The committees of the Parliament then invite the Ministries to explain and discuss their Budget Framework Papers. The committees can hold public hearings, but that does not seem to be usually the case. The Natural Resources Committee has not yet held hearings in the budget process, even though it organizes hearings for other purposes. Parliamentarians are invited to events organized by NGOs who are engaged in budget issues. The Parliament provides then feed-back to MFPED, which tries to accommodate requests as far as possible. Subsequently, the Ministry provides feedback to the Parliament to explain which requests were addressed and which ones could not be addressed and why. Typical reasons for not incorporating requests are resource scarcity, or program design. Requests are often referred to the subsequent year. At the Parliamentary level, there is one budget officer for one or for two committees to assist the Parliamentarians in the scrutinizing the details of the budget. The Parliament also has a Research Department, which has, however, limited capacity for assisting Committees. Budget Speech: On June 14, the Budget Speech is delivered. This day is fixed for all countries of the East African Community. Deliberations in Parliament and Appropriation of the Budget: After the Budget Speech, the Accountant prepares the documents for the closing of the financial year. The Ministries work out the Policy Statements for their Sector Budgets, which are submitted to Parliament, which scrutinizes and discusses them. The Sector Committees of the Parliament then decide on changes, typically in consensus with the Cabinet, before the Budget is finally appropriated, usually between mid to end of August. Opportunities and challenges of the budgetpreparation process 6. As compared to other African countries, Uganda's budget process is positively special in several regards. The process is designed to ensure coordination among different Ministries and Agencies that share mandates for the same sector, which is important for SLM issues. Moreover, the process is aimed to provide transparency from early stages of the budget preparation process onwards, and to involve stakeholders from the private sector and civil society in the process of budget preparation. In fact, considerable resources are spent to organize budget conferences with wide participation at the national, regional and local levels. The Sector Working Groups also provide room for private sector and civil society participation. Through the Budget Act the Parliament has succeeded in being formally involved in the budget preparation process. Moreover, the Parliament has resources, in form of the Budget Office, to assist Parliamentarians in scrutinizing the budget. A vibrant civil society takes an active role in the budget process, including by organizing their own events related to the budget, and by conducting their own analyses of the budget, as in case of the Uganda Debt Alliance (Renzio et al., 2006). NGOs that deal with sustainable land management issues, such as the Uganda Land Alliance, have engaged in the budget process. 7. In spite of these opportunities, major challenges regarding the budget preparation process remain. They include: (i) the declining discipline of keeping time- lines; (ii) the role of indicative budget ceilings and budget allocation processes; (iii) the limited information of "value-for-money" in the Budget Framework Papers; and (iv) the limited reflection of priorities of civil society and farmer organizations in budget decisions. Timing of the process: Even though the time line of the budget process is well specified in the Budget Act, delays frequently happen, which limit the possibilities for meaningfhl consultations and deliberations during the budget process. It seems that the discipline of keeping time-lines has been deteriorating over time, which may indicate a mismatch between the requirements of the process, and the capacity and incentives to follow it on a sustained basis. Role of indicative ceilings and budget allocation process: The indicative budget ceilings declared by MFPED is considered as a major constraint to exercising discretion in the budget preparation process. Representatives of Ministries, semi- autonomous agencies and the Parliament alike stated that adjustments can only take place within these ceilings, and that there is hardly any possibility to increase the budget allocationbeyond the ceilings, at least in the short run. The ceilings are set by a decision of the Cabinet, and MFPED ensures that the budget ceilings are aligned with strategic planning priorities, as specified in the PEAP and the Sector Development Strategy and Investment Plans. However, unless there are well- specified and agreed criteria on how to decide whether proposed budget allocations reflect the priorities specified in planning documents, there is obviously considerable room for discretion. Sector Investment Plans, which specify resource allocations for priorities, can help to address this problem. Therefore, to receive appropriate budget allocations, it is important to make sure that SLM issues are adequately addressed in the National Development Plan currently under development, and in all Sector Investment Plans concerned. Quality of the Budget Framework Papers: In view of their central role in the budget process, the quality of the Budget Framework Papers deserves particular attention. The Agriculture Sector Public Expenditure Review identified a number of issues that could be improved (World Bank, 2008), most notably a better match between the budget proposals in the Framework Paper and the proposals in the Development Strategy and Investment Plan. At least, the often rather large differences would need to be explained to provide transparency and avoid a mismatch between planning and budgeting. In the Budget Framework Papers' section on past performance, some sectors compare outcomes against targets, which allows for an assessment of achievements. This is not the case for the agricultural sector. There is limited value in information such as "various rainfall, soil fertilitylproductivity improvement and conservation practices were demonstrated to farmers and other stakeholders visiting the Institute", to mention an SLM-related example. A more meaninghl indicator would be adoption rates of such techniques. With the exception of one statement about increased in milk yield and dry season crop margin, all past performance achievements in the agricultural sector are reported as outputs, not outcomes. Likewise, in the Environment Sub-sector, only one item (60 percent of entire Central Forest Reserve acreage free from encroachment) specified outcomes. Likewise, the performance indicators for planned spending have the form of outputs rather than outcomes. Therefore, the Budget Framework Papers provide rather limited information on the effectivenessof investment. Civil society participation: In interviews held for this review, civil society members expressed the concern that they are only invited to comment on the drafts of the Budget Framework Papers that have already been developed, rather than taking a meaningful part in the development of the Framework Papers. Moreover, they felt that their suggestions were often not taken into consideration at all, and that "token participation" has become a widespreadproblem. It has also been criticized that civil societies' participation in the budget process is by invitation, therefore not all interested organizations can in fact participate (Renzio et al., 2006). The interviewed representatives of the Uganda National Farmers' Federation explained that their organization is not involved in the budget process. It also appears that the capacity of civil society groups to analyze the budget with regard to poverty and gender issues is more advanced than their capacity to analyze the budget with regard to environmental and SLM issues. 1.2 Execution of the Budget 8. Uganda has a cash budget system, according to which disbursements are adjusted to the available resources. Frequently, the resources are less than projected due to uncertainties in projections and due to emergencies, such as flooding. According to MFPED, there is no contingency budget to account for such events. Therefore, the budget allocations to Ministries and Agencies are reduced according to the reduced availabilityof resources. 9. However, budget items that fal,l under the Poverty Action Fund (PAF) are protected up to a level of 95 percent. The reason is that PAF funds are based on earmarked donor budget support and on funds of the Highly Indebted Poor Country (HIPC) debt relief initiative. NAADS funding is in the PAF category, but most of the SLM activities do not fall under PAF. 10. The fact that some budget items are protected leads to even stronger fluctuations in those budget items that are not protected. If the budget allocation to a sector is reduced, it is typically not possible to cut all budget items equally in proportion to the reduced availability of fimding. Some budget lines such as salaries for staff cannot be reduced. In some cases, such as new projects, absorptive capacity can also be a problem, since it often takes some time until project procedures can be implemented smoothly. 11. A major mechanism for monitoring and evaluating budget execution is the Public Expenditure Review. Six months after the appropriation of the budget, MFPED conducts a Public ExpenditureReview, which covers the previous fiscal year and the first 6 months of the current year. The national review is based on annual sector reviews, which are typically conducted asjoint sector reviews, involving Government, donors and civil society. Parliamentarians also participate in such reviews, often as discussants in conferenceswhere the reviews are presented. In these reviews, performance is compared to the Sector Investment Plans. 12. Although the Public Expenditure Review constitutes a potentially important mechanism to provide feed-back from past performance for future budget planning, the performance there has been measured by outputs, not by outcomes. This limits their application to change priorities in budget preparation. With regard to SLM, the availability of data on outcomes, and the possibility to link outcomes with public spending is a challenging and data-intensive task. A stronger involvement of research institutes that are specialized on SLM issues might be useful to meet this challenge. Likewise, the Sector Working Groups may play a stronger role in effectively monitoring budget execution (World Bank, 2008). 13. With regard to financial accountability in budget execution, Uganda's Public Finance and Accountability Act of 2003 (PFAA) is widely considered to provide a strong legal accountability framework at both central and sub-national levels (World Bank, 2004). The Act specifies the authority and obligations of the Accountant General in the management of treasury and accounting functions, as well as that of the Accounting Officer of a ministry or agency. In each Ministry or self-accounting agency, the Accounting Officer, typically the Permanent Secretary or head of the agency, is personally responsible for the budget of hisher organization. The institutions in charge of controlling public spending at the national level also include Office of the Auditor General, the Inspectorate of Government, and the Public Accounts Committee of the Parliament. The Office of the Auditor General is an independent authority with the mandate to provide oversight of government operations through financial and other management audits. The Inspectorate of Government also plays an important role in promoting accountability, and is in charge of monitoring the PAF funding, which includes some SLM-related activities (e.g. agriculture, water, and sanitation). The Inspectoratehas published two National Integrity Surveys. 14. This description shows that the legal framework for controlling corruption is comparatively strong, and that institutions and processes for accountability are in place. Problems may exist, however, regarding the institutional capacity and the incentives to use existing mechanisms of accountability effectively. The availability of recent representative data on the hctioning of accountability mechanisms for SLM activities is limited. The last National Integrity Survey published by the Inspectorate of Government in 2003 covered the services provided by the Agriculture Department, the Veterinary Department and the Water Department, which are related to SLM. Less than 8 percent of the interviewed households considered these Departments as "extremely or largely corrupt", which compares favorably with Health Units (rated by 32 percent of the households to be extremely or largely corrupt) or Primary Schools (rated by 13percent to be in this category) (Inspectorateof Government,2003). 2. TheBudgetprocesses at the local level 2.1. Budgetpreparation Theprocess 15. The budget process at the District level is integrated with the budget process at the national level. To promote participatoryplanning, the Ministry of Local Government has developed a "Harmonized Participatory Planning Guide for Lower Local Governments" (Republic of Uganda, 2003). At the District visited for this study, the budget process was described as follows: Needs Assessment: This should take place from October to December. Civil society organizations, opinion leaders and church leaders are expected to participate in this process. It involves meetings and planning activities at the sub- county level, the parish level and the village level. At the village level, a situation analysis should be conducted, involving data collection and the preparation of a village map. Under the guidance of a technical officer, villagers should prioritize activities and develop Village Action Plans. The Village Action Plans are combined to Parish Action Plans. On this basis, Sub-county Development Plans are prepared. These plans feed into the District Budget Framework Paper. Budget Conference: To provide feedback on the District Budget Framework Paper, a Budget Conference is held, typically in December. The Conference is attended by approximately 150 people, including community leaders, political leaders and technical staff. Approval of the Budget: The District Budget Framework Paper is subject to review by the District Technical Planning Committee (DTCE), the District Executive Committee, formed by the District Chair and the Secretaries (District Council Members that have the status of local ministers), and the committees of the District Council, before it is approved by the District Council itself. Likewise, the budgets of the Sub-counties are reviewed by the Sub-county Technical Staff, the Sub-county Executive, and the committees of the Sub-county Council prior to approval by the Sub-county Council. 16. The Districts receive indicative budget planning figures from the central-level Ministries at Regional Budget Workshops. This workshop should ideally be held in October, but in practice are often held in December. Due to the information distributed from the central level at the workshops, the heads of the different Directorates at the District level can make their plans, taking the availability of finds from national-level programs into account. If the heads of Directorates plan activities, they have to indicate the source of h d i n g for these activities. If no funding from central programs is available, they can suggest finding from the Districts' own resources, but the likelihood to get h d i n g from these sources is small. Most of the resources that the Districts generate themselves are used for administrative purposes. Table 9 displays the Mukono District Budget as an example. Table 9: Structure of Mukono District budget, in percent of total Administration 8.0 Council Commissions and Boards (e.g. Land Board) 1.4 Finance and planning 11.2 Production and extension 6.6 Health 14.5 Education 46.6 I Technical services and works 10.9 Natural resources 0.3 Community-basedservices (culture, gender issues, etc.) 0.5 Total 100.0 Source:Mukono Districtrecords (2007) 17. There is limited discretion by districts over budget resources to adopt SLM measures at local level. As can be seen from Table 9, the allocation to agriculture (referred to as production and extension) was 6.6 percent of the budget, while the allocation to natural resources was 0.3 percent, the lowest of all budget shares. Approximately 40 percent of the budget for production and extension is allocated to NAADS, but NAADS budget is determinedby the national government. 18. The planning process for NAADS is somewhat different from that of other agencies as it aims at empowering farmers' organizations. NAADS has created a structure of local level farmers' groups, which elect Farmers' Fora at Sub-county and District levels. The farmers' groups at the village level identify priority commodities that the advisory service should focus on. From these proposals, the Farmers' Forum at the Sub-county level selects three priority commodities. An annual work plan is developed for these enterprises. The activities are budgeted, based on the indicative budget figures provided by the central government. The Sub-county Farmers' Forum consists of six members, three of which form the Procurement committee and three the Executive Committee. The Procurement Committee is joined by the NAADS Sub-county Coordinator and the Sub-county Secretary for Production (a member of the Sub-county Council) as co-opted members. The Committee invites Expressions of Interests from companies who want to provide advisory services, and then invites one applicant per commodity to prepare a Technical and Financial Proposal (earlier, two proposals were invited). A contract is awarded after negotiations. According to the NAADS guidelines, proposals have to specify how environment is addressed in the work to be conducted, next to HNIAIDS, gender mainstreaming and poverty targeting. The role of the guidelines is hrther discussed below. Opportunities and challenges 19. The opportunities and challenges of the budget preparation process are rather similar to those encountered at the national level. The fact that the budget process is designed to be participatory is certainly an opportunity. With regard to political, administrative and fiscal decentralization, Uganda is among the most advanced countries in Sub-Saharan Africa (Ndegwa, 2000). Moreover, far-reaching provisions have been made for bottom-up planning. However, important challengesremain: Timing of the process: Even though the timeline for the budget preparation process is well-specified because the process is integrated with the national budget process, the ability to keep the timeline has apparently declined over time. This limits the possibilities to perform all steps of the budget preparation process as specified in the respective guidelines. At the local level, this problem is exaggerated by limited capacity compared to the large number of activities to be performed at various levels of local government. Complexity of the process: Moreover, the planning procedures for bottom-up planning are extremely complex. The "Harmonized Participatory Planning Guide for' Lower Local Governments" specifies eighteen different steps, including SWOT (Strength, Weaknesses, Opportunities, and Threats) analysis, and visioning exercises. A review conducted by the DCTE in the District visited for this study revealed that many villages do not have an action plan due to the lack of sufficient staff and resources in the local government administration to facilitate the prescribed planning processes. Moreover, the District Development Plan, which should be developed in advance of the budget, is often developed in parallel due to time constraints. Limited discretion over the use of budgetary resources: Partly, this problem is caused by the low capacity of districts to raise their own resources. Mukono District visited for the study has a total budget of about UGX 33 billion. The own revenues generated by the District are in the range of UGX 400 million, which indicate that even a relatively prosperous District close to the Nation's capital can have a low capacity to raise own revenues (only 1percent of the budget). With regard to the budgetary resources received from the centre, the assessment of the interviewed DCTE members is that they can have discretion to make their own decisions for approximately 40 percent of these resources, while 60 percent are earmarked for specific programs. The problems associated with the indicative ceilings for these resources are similar to those described above for the national level. Limited influence of communities: Table 10 displays the perceived influence of the different actors on the final allocation of the District Budget in Mukono. The perceived influence was derived fiom the application.of a mapping technique, which helps to visualize the respondents' perception on the influence of different stakeholders on a specified outcome (Schiffer, 2007). As shown in Table 10, the donor-funded national programs, which are implemented through the Districts, have the largest influence on the District budget. However, as indicated above, they leave relatively little discretion to the Districts. Table 10 also shows that the political heads of the Districts and Sub-counties have a comparatively large influence on the budget. In spite of the efforts to implement a participatory bottom-up planning process, the village committees are not perceived to have any real influence on the final District Budget, probably due to the implementation challenges discussed above. Table 10: Perceived influence of different actors on district and sub-county budget (Mukono District) I Parliamentarian from the District I 3 I 1 District Chair and Executive Committee I 4 I District Council and its Committees 2 District Technical Planning Committee and District Directorates 3 Sub-county Council and its Committees 2 Sub-county Executive 4 Public administration staff at sub-countv level 2 I Civil societv oreanizations I 1 I [ Village Committee 0 Source: Focus group discussion with District Technical Planning Committee (2007). Low priority for environmental matters: According to the experience of the interviewed DCTE members, environmental matters receive rather low priority in the planning process. At the village level, community members are not usually willing to dedicate public resources for natural resource management activities, considering that other needs are considered to be more pressing. As shown in Table 9, almost half of the District budget is spent on education, while natural resource management receives only 0.3 percent of the budget. The District has prepared an Environmental Action Plan, but it is obviously difficult to implement this plan given the limited amount of funding. Moreover, as the interviewed DCTE members observed, the planning and budgeting guidelines aim at ensuring the provision of infrastructure and services that can be used by the entire community. SLM activities that target groups of farmers or individual farmers cannot be financed under the general budget. Only the NAADS budget would allow for this approach. 2.2. Budget Execution 20. Similarly to the situation described above for the national level, budget execution at the District level is characterized by fluctuating resource availability. Since PAF funding is protected, budget line items that fall under PAF, such as the NAADS program, are not subject to budget cuts, which results in disproportionately larger cuts in other budget items. 21. As in the case of the national level, the budget process at the District level includes a review of previous year's performance. In principle, this review should be conducted in December at the beginning of the budget cycle. The review is done on the basis of quarterly progress reports. The reference for assessing performance and progress is the District Development Plan, which is a well-elaborated planning document for a three-year period. Similarly to the national level, this review is potentially a strong instrument to guarantee feed-back. However, due to a focus on outputs rather than outcomes, and due to the challenges of collecting data on SLM, the actual possibilities to use information on the effectiveness of public spending for planning purposes remain limited. 22. The institutions for accountability mentioned above, the Auditor General and the Inspectorate of Government are also in charge of Local Governments. A recent report by the Inspectorate reveals a number of problems in accountability at the local government, such as irregularities in procurement (Inspectorate of Government, 2006). The Ministry of Local Government (MOLG) is in charge of supporting the local governments, which play a key role in Uganda's decentralized system. MOLG has a Local Government Inspectorate Department, which-among other responsibilities- analyzes the financial reports of local governments. Districts also perform an internal audit. According to the National Integrity Survey of 2003, 11percent of the sample households considered Village Councils "extremely or largely corrupt", and 12percent rated Sub-county Councils to be in this category (Inspectorate of Government, 2003). These figures are somewhat higher than the ratings for the Agriculture, Veterinary and Water Departments quoted above. References Balihuta, A. and K. Sen (2001): Macro-Economic Policies and Rural Livelihood Diversity: A Ugandan Case Study. LADDER Working Paper 3, University of East Anglia, U.K. Bashaasha, B. (2001): The Evolution and Characteristics of Farming Systems in Uganda. Kampala: Makerere University. Mimeo. Bazaara, N. (2003): Decentralization, Politics and Environment in Uganda. Washington, D.C.: World Resources Institute (WRI), Governance and Institutions Program. Benin, S., E. Nkonya, G. Okecho, J. Pender, S. Nahdy, S. Mugarura, E. Kato and G. Kayobyo (2007): Assessing the Impact of the National Agricultural Advisory Services (NAADS) in Uganda Rural Livelihoods. Washington, D.C.: IFPRI Discussion Paper 724. Bojo, J. (1996): The Cost of Land Degradation in Sub-Saharan Afiica. Ecological Economics 16: 161-173. Boserup, E. (1965): The Condition of Agricultural Growth. New York: Aldine Publishing. Chenerey, E. (1960): An Introduction to the Soils of the Uganda Protectorate. Mem. Res. Div. Dep. Agric. Uganda, Series 1,No.1. Chomitz, K.M. (2007): At Loggerheads? Agricultural Expansion, Poverty Reduction and Environment in the Tropical Forests. Washington, D. C.: The World Bank. Clay, D., V. Kelly, E. Mpyisi and T. Reardon (2002): Input Use and Conservation Investments among Farm Households in Rwanda: Patterns and Determinants. Natural Resources Management in African Agriculture. C. Barrett, F. Place and A. Aboud. New York, CABI. Cline, W. (2007): Global Warming and Agriculture. Impact Estimate by Country. Washington, D.C.: Center for Global Development and Peterson Institute for International Economics. Defoer, T., A. Budelman, C. Toulmin and S. Carter (2000): Building Common Knowledge: Participatory Learning and Action Research. Managing Soil Fertility in the Tropics: A Resource Guide for Participatory Learning and Action Research. T. Defoer and A. Budelman. Amsterdam, the Netherlands: Royal Tropical Institute (KIT) 1, 39-66. Deininger, K. and R. Castagnini (2004): Incidence and Impact of Land Conflict in Uganda. Washington, D.C.: World Bank Policy Research Working Paper 3248. Deininger, K. and J. Okidi (2001): Rural Households: Incomes, Productivity, and Nonfarm Enterprises. Uganda's Recovery: The Role of Farms, Firms, and Government. R. Reinikka and P. Collier. Washington, D.C., The World Bank. Dregne, H. (1990): Erosion and Soil Productivity in Africa. Journal of Soil and Water Conservation45: 432-436. Dreschel, P., L. Gyiele, D. Kunze and 0 . Cofi (2001): Population density, Soil Nutrient Depletion, and Economic Growth in Sub-SaharanAfrica. Ecological Economics 38, Issue 2: 251-258. Feder, G., T. Onchan, Y. Chalamwong and C. Hongladaron (1988): Land Policies and Farm Productivity in Thailand. Baltimore: The Johns Hopkins University Press. GEF (2006): Strategic Investment Program for Sustainable Land Management in Sub- Saharan Africa. Draft. Global Environment Facility. GoU (2001): National Biomass Energy Demand Strategy 2001-2010. Kampala: Ministry of Energyand Mineral Development,Uganda. Henao, J. and C. Baanante (2006): Agricultural Production and Soil Nutrient Mining in Africa: Implications for Resource Conservation and Policy Development. IFDC, Muscle Shoals, AL. Holden, S. and B. Shiferaw (2002): Poverty and Land Degradation: Peasants' Willingness to Pay to Sustain Land Productivity. Natural Resources Management in African Agriculture. C. Barret, F. Place and A. Aboud. New York, CABI. Holden, S., B. Shiferaw and M. Wik (1998): Poverty, Credit Constraints and Time Preference of Relevance for Environmental Policy. Environment and Development Economics 3: 105-130. IFDC (2001): The Fertilizer Market Assessment and a Strategy for Development. Kampala: International Fertilizer Development Corporation. Report submitted to Sasakawa-Global2000 and USAID, Kampala Uganda. Inspectorate of Government (2006): Report to Parliament January 2006 to July 2006. - Kampala: Republic of Uganda. Inspectorate of Government (2003): Second National Integrity Survey - Final Report. Kampala: Inspectorateof Government, Uganda. Jagger, P. and J. Pender (2003): Impacts of Programs and Organizations on the Adoption of Sustainable Land Management Technologies in Uganda. Washington, D.C.: InternationalFood Policy Research Institute(IFPRI), EPTD Discussion Paper 101. Jayne, T., V. Kelly and E. Crawford (2003): Fertilizer Consumption Trends in Sub- Saharan AErica. USAID, Office of Sustainable Development, Africa Bureau, Policy Synthesis69. Kaizzi, K., J. Byalebeka, C. Wortrnann and M. Mamo (2007): Low Input Approaches for Soil Fertility Management in Semiarid Eastern Uganda. Agronomy Journal 99: 847-853. Karugaba, P. (2001): Big Tobacco's Overseas Expansion: Focus on Uganda. The Environmental Action Network for the San Francisco Tobacco Free Project. UICC GLOBALink. httD://www.alobalink.ordtobacco/docs/af-docs/01O5karunaba.shtml. Koehlin, G. (2006): Aspects of Land Degradation in Lagging Regions: Extent, Driving Forces, Responses and Further Research with Spatial References to Ethiopia. Paper presented at the World Development Report Agriculture and Rural Development InternationalPolicy Workshop. September 4,2007, Berlin. LaFrance, J. (1992): Do Increased Commodity Price Lead to More or Less Soil Degradation? Australian Journal of Agricultural Economics 3691: 157-182. Larson, B. and G. Frisvold (1996): Fertilizers to Support Agricultural Development in Sub-SaharanAfrica: What is Needed and Why. Food Policy 212(6): 509-525. Lind, J. and J. Cappon (2001): Realities or Rhetoric? Revisiting the Decentralization of Natural Resource Managementin Uganda and Zambia.Acts Press Nairobi, Kenya: 5-72. Lindblade, K., J. Turnuhairwe and G. Carswell (1996): More People, More Fallow: Environmentally Favorable Land-Use Changes in southwestern Uganda. Report for the Rockefeller Foundation and CARE International. MAAIF (2008): Ministry of Agriculture, Animal Industry and Fisheries: http://~~~.ag;riculture.go.ug/statistics.htrn. MAAIF and MFPED (2000): Plan for Modernization of Agriculture (PMA): Eradicating Poverty in Uganda, GovernmentPrinter. Kampala: Uganda. Magunda, M. and M. Majaliwa (2006): Soil Erosion in Uganda : A Review. Kampala: Nile Basin InitiativeIssuePaper, Uganda. Magunda, M. and M. Tenywa (1999): Soil and Water Conservation. Kampala: Mimeo, Uganda National AgriculturalResearch Organization (NARO), Uganda. McIntire, J., D. Bourzat and P. Pingali (1992): Crop-Livestock Interaction in Sub- Saharan Africa. Washington, D.C: The World Bank. MNR (1992): The National Environmental Action Plan for Uganda. Kampala: Ministry of Natural Resources,NEAP Secretariat,Uganda. Morris, M., V. Kelly, R. Kopicki and D. Byerlee (2007): Fertilizer Use in Afiican Agriculture. Lessons Learned and Good Practice Guidelines. Washington, D.C.: The World Bank. Mugisha, S. (2002): Patterns and Root Causes of Land Cover/Use Change in Uganda: An Account of the Past 100 Years. Nairobi: LUCID Working Paper 14, International LivestockResearch Institute (ILRI), Kenya. Mukherjee, C., H. White and M. Wuyts (1998): Econometric and Data Analysis for Developing Countries.London: Routledge. MWLE (2002): The National Forest Plan. Kampala: Ministry of Water, Lands, and Environment, Uganda. MWLE (2001): The Uganda Forest Policy. Kampala: Ministry of Water, Lands, and Environment, Uganda. Ndegwa, S. (2000): Decentralization in Afiica: A StocktakingSurvey. Washington, D.C.: Afiica Region Working Paper Series40. World Bank. NEMA (2007): Mainstreaming Environmental Issues into Budget Framework Papers - Users' Manual. Kampala: National Environment ManagementAuthority. NEMA (2005): State of the Environment, 2005106. Kampala: National Environmental ManagementAuthority, Uganda. NEMA (2001): State of the Environment, 2000101. Kampala: National Environmental Management Authority, Uganda. NEMA (1998): Stateof the Environment, 1997198.Kampala: NEMA, Uganda. NEPAD and FA0 (2004): National Medium-Term Investment Programme. Support to NEPAD-CAADP Implementation. Online at: http:l/ftr,.fao.ornldocre~lfaol007/ae560elae560e00.pdf. Nkonya, E., J. Pender, C. Kaizzi, E. Kato, S. Mugarura, H. Ssali and J. Muwonge (2007): Land Management and its Linkages to Poverty in Sub-Saharan Africa: The Case of Uganda. FPRI Mimeo. Nkonya, E., P. Gicheru, J. Woelcke, B. Okoba, D. Kilambya and L. Gachimbi (2006): Out of Sight, Out of Mind: Quantifying the Long-Term Off-Site Economic Impacts of Land Degradation in Kenya. Paper Prepared for Presentation at the American AgriculturalEconomics Meeting, Long Beach California, July 23-26,2006. Nkonya, E., C. Kaizzi and J. Pender (2005a): Determinants of Nutrient Balances in a Maize Farming System in EasternUganda. Agricultural Systems 85: 155-182. Nkonya, E., J. Pender, C. Kaizzi, E. Kato and S. Mugarura (2005b): Policy Options for Increasing Crop Productivity and Reducing Soil Nutrient Depletion and Poverty in Uganda. Washington, D.C.: FPRI Environment and Production Technology Division Paper 134. Nkonya, E., J. Pender, P. Jagger, D. Sserunkuuma, C. Kaizzi and H. Ssali (2004): Strategies for Sustainable Land Management and Poverty Reduction in Uganda. Washington, D.C.: InternationalFood Policy ResearchInstitute, Research Report 133. Pender, J., S. Ssewanyana,E. Kato and E. Nkonya (2004): Linkages between Poverty and Land Management in Rural Uganda: Evidence from Uganda National Household Survey, 1999100. Washington, D.C.: International Food Policy Research Institute, Environment and ProductionTechnologyDivision Discussion Paper 122. Pender, J., P. Jagger, E. Nkonya and D. Sserunkuuma (2001): Development Pathways and Land Management in Uganda: Causes and Implications. Washington, D.C.: InternationalFood Policy Research Institute, EPTD Discussion Paper 85. Pender, J. and J. Kerr (1999): The Effects of Land Sales Restrictions: Evidence fiom SouthIndia.Agricultural Economics 21: 279-294. Pender, J. (1996): Discount Rates and Credit Markets: Theory and Evidence from Rural India.Journal of Development Economics 50: 257-296. Place, F., R. Masupayi and K. Otsuka (2001): Tree and Cropland Management in Malawi. Land Tenure and Natural Resource Management: A Comparative Study of Agrarian Communities in Asia and Africa. K. Otsuka and F. Place. Baltimore, Johns Hopkins. Place, F. and P. Hazel1 (1993): Productivity Effects of Indigenous Land Tenure Systems in Sub-Saharan Africa.American Journal of Agricultural Economics 75(1): 10-19. Reardon, T. and S. Vosti (1995): Links between Rural Poverty and the Environment in Developing Countries: Asset Categories and Investment Poverty. World Development 23(9): 1495-1506. Renard, K., G. Foster, G. Weesies and J. Porter (1991): RUSLE: Revised Universal Soil Loss Equation.Journal of Soil and WaterConservation46(1): 30-33. Renzio, P., V. Azeem and V. Ramkurnar (2006): Budget Monitoring as an Advocacy Tool - Uganda Debt Network. http://www.internationalbudnet.org/ug;anda-UDN.pdf: The InternationalBudget Project. Republic of Uganda (2007): Climate Change - Uganda National Adaptation Programs of Action. Kampala, Uganda: Republic of Uganda (2003): Harmonized Participatory Planning Guide for Lower Local Governments. Kampala:Ministry of Local Government, Uganda. Republic of Uganda (1998): Land Act, 1998. Government Printer. Kampala:Uganda. Ruben, R., J. Pender and A. Kuyvehnhoven (2007): Sustainable Poverty Reduction in Less-Favored Areas: Problems, Options, and Strategies.SustainablePovertyReduction in Less-Favored Areas. R. Ruben, J. Pender and A. Kuyvehnhoven, CABI, Wallington U.K.: p. 1-61. Schiffer, E. (2007): The Power Mapping Tool - A Method for the Empirical Research of Power Relations. Washington, D.C.:Discussion Paper No. 703, InternationalFood Policy Research Institute (IFPRI). Slade, G. and K. Weitz (1991): Uganda Environmental Issues and Options. A Masters Dissertation.Duke University,North Carolina,USA. Smaling, E., J. Stoorvogeland P. Windrneijer (1993): Calculating Soil Nutrient Balances in Africa at Different Scales.District Scale. FertilityResearch, 35: 237-250. Ssali, H. (2002): Soil Organic Matter in Uganda and its Relationship to Major Farming Systems. Washington, D.C.: Resource Paper Submitted to the International Food Policy Research Institute. Sserunkuurna, D., J. Pender and E. Nkonya (2001): Land Management in Uganda: Characterization of Problems and Hypotheses about Causes and Strategies for Improvement. Washington, D.C.: InternationalFood Policy Research Institute,Mimeo. TerrAfrica (2007): Production of Financing Guidelines: Financing Issues for Scaling Up SLM: A ResourceGuide for PolicyMakers. Draft, October 5,2007. Washington, D.C.: TerrAfrica(2005): Country Engagement Strategy. Tiffen, M. (2003): Transition in Sub-Saharan Africa: Agriculture, Urbanization, and Income Growth. WorldDevelopment 31.8: 1343-1366. UBOS (2006): Uganda National Household Survey 2005/06. Socio-Economic Report. Entebbe: Uganda Bureau of Statistics. UBOS (2001): Uganda National Household Survey 1999/2000. Report on the CommunitySurvey. Entebbe: Uganda Bureau of Statistics. Uganda, R. o. (2001): The Budget Act. Kampala: Republic of Uganda. Voortman,R., B. Sonneveld and M. Keyzer (2000): African Land Ecology: Opportunities and Constraints for Agricultural Development. Center for International Development (CID), Harvard University. Working Paper 37. WDR (2007): World Development Report 2008: Agriculture for Development. Washington, D. C.: The World Bank. Woelcke, J. (2006): Technological and Policy Options for Sustainable Agricultural Intensificationin Eastern Uganda. Agricultural Economics 34(2): 129-139. Woelcke, J. (2003): Bio-Economics of Sustainable Land Management in Uganda. Development Economics and Policy. Vol. 37.: Peter Land Europaeischer Verlag der Wisssenschafien. World Bank (2008): Uganda: Agriculture Sector Public Expenditure Review. Kampala: Ministry of Agriculture,Animal Industries and Fisheries of Uganda; Ministry of Finance, Planning and EconomicDevelopmentof Uganda; World Bank; DFID; ReSAKSS (ILRI). World Bank (2006): Sustainable Land Management: Challenges, Opportunities, and Trade-offs. Washington, D. C.: The World Bank. World Bank (2004): The Republic of Uganda: Tracking Poverty-Reducing Spending: Second Assessment and Action Plan. Washington, D.C.: http://www1.worldbank.or~~ublicsector/ve/HIPC/UgandaAAP.~df: World Bank. Wortmann, C. and C. Kaizzi (1998): Nutrient Balances and Expected Effects of Alternative Practices in Farming Systems of Uganda. Agriculture, Ecosystems and Environment 71: 115-129. WRI (2007): Forests, Grasslands and Drylands - Uganda Database. World Resource Institute,online at: http://earthtrends/wri.org/. Zake, J., M. Magunda and C. Nkwiine (1997): Integrated Soil Management for Sustainable Agriculture and Food Security. The Ugandan Case. Paper presented at the FA0 workshop on Integrated Soil Management for Sustainable Agriculture and Food Securityin Southern and Eastern Africa, Harare, Zimbabwe on December 8-12, 1997.