Understanding land dynamics and livelihoods in refugee hosting districts of Northern a

Understanding Land Dynamics and Livelihood in Refugee Hosting Districts of Northern Uganda Understanding land dynamics and livelihoods in refugee hosting districts of Northern Uganda c

©Copyright 2018

United Nations Development Programme (UNDP) Plot 11, Yusuf Lule Road P.O. Box 7184, , Uganda Tel: +256 417112100 Understanding land dynamics and livelihoods in refugee hosting districts of Northern i Uganda

Foreword

While the rest of the world is still reluctant to fully for production is at sub-optimal level to assure integrate refugees, Uganda has, for several food self-sufficiency as well as a sustainable decades, recognized the need for a shift from livelihood. Considering the increased demand policies that kept refugees confined to camps. for land in the foreseeable future, guaranteeing Applauded as one of the most progressive in official land rights to host communities can be the world, Uganda’s refugee policy profoundly a viable assurance mechanism to facilitate commits to enabling refugees to pursue self- acquisition of land by Government as well as reliance and a dignified life while in exile, not only to enable ease of transactions between hosts out of compassion, but also to empower them and refugee communities. It will ensure trust to contribute to their new host communities, as and facilitate the envisaged productive use workers, tax-payers and as consumers. of land. Second, in addition to enhancing the optimal utilization of land for production through During the Solidarity Summit on Refugees held in improved technologies, there is an urgent need 2016, Uganda was recognised for its generosity to implement interventions that will diversify in hosting nearly 1.3 million refugees, at the time. livelihood options by involving the private The number has since increased to 1.4 million. sector through developing effective incentives The international community was called upon specifically targeting refugee hosting districts. to emulate the country’s generosity. Emulating Third, recognizing the opportunity presented Uganda’s approach is novel because in the due to the transformation of villages to small face of the daunting challenges of protracted rural towns due to the influx of refugees, it is displacement, waning humanitarian assistance, imperative to empower Local Governments in and the inclusivity of the 2030 Agenda, models refugee management, to not only guarantee that emphasise assisting refugees to “assist community cohesion, but also coherence in themselves’ must take center stage. local development and physical planning. Not least of all, the environment, if in grave danger, then people and the future are equally In addition to the myriad services that Ugandans endangered. We should galvanise efforts to extend to refugees, including education, revamp this natural capital, but also devise healthcare, water, security and protection, mechanisms for sustainably utilising it. refugees receive land for settlement and agricultural use to supplement relief support. While it is known that refugees receive land, I thank my staff for taking lead in the preparation information on modalities of access, use, of this report. I appreciate the unwavering productivity and community relations regarding support of the Office of the Prime Minister that this productive asset is still largely anecdotal. This guided preparation of this report, and the informed my decision to commission this report Economic Policy Research Centre that led the to contribute to the growing knowledge on research process. UNDP remains steadfast in refugee management in Uganda. I am glad that its commitment to support the Government the report addresses important aspects, while of Uganda and other actors in their quest elucidating the growing interconnectedness to mitigate risk and sustain the livelihoods of between refugees and their hosts, which I find vulnerable refugees and their host communities critical to sustaining co-existence.

The report underscores several critical interventions to sustain Uganda’s progressive Rosa Malango model, which I implore policy makers and relevant development practitioners to consider critically in their programming. The findings of the report show that while land is used effectively UN Resident Coordinator & by refugees for settlements, the use of land UNDP Resident Representative Understanding land dynamics and livelihoods in refugee hosting districts of Northern Uganda ii

Acknowledgment

This report was commissioned and coordinated by The United Nations Development Programme (UNDP) Uganda, under the supervision of the UNDP’s Strategy and Policy Unit led by Yemesrach Workie (Economics Advisor) with Support from Tony Muhumuza, Simon Peter Nsereko, and Josephine Faith Nansubuga. It was prepared by the Economic Policy Research Centre (EPRC) team comprised of Gemma Ahaibwe, Sheila Depio and Miriam Katunze. The Office of the Prime Minister including Hon. Gerald Menya (Commissioner for Refugees), Titus Jogo, Solomon Osan, Robert Baryamwesiga, David Wangwe, and Darlson Kusasira provided invaluable guidance during the preparation of the report. UNDP also appreciates support from participants of the consultative meetings including Francesca Akello (UNDP), Wilfred Bibanga (FAO), Francesca Cozzarini (UNDP), Mugero Jesse (Refugee Law Project), Wilson Kwamya (UNDP), Lucy Ndungu (UNDP), Lilian Otiego (UNHCR), and Ranya Sherif (UNHCR). Valuable insights were also given by Rose Twine (Eco Stove), Edoni Brahan Nasur (DanChurch Aid), Florence Ochola (UNDP) and Joan Teria (UNDP). The report benefited from wide ranging consultations held in Kampala and selected refugee hosting districts. Key informants were drawn from settlement commandants, Refugee Welfare Officers, Refugee Desk officers, Local Government officials, cultural leaders, clan heads and NGOs in the respective districts. We acknowledge the valuable inputs from members of Focused Group Discussions in the study districts that enriched the report. We duly acknowledge the Food and Agriculture Organisation (FAO) for sharing the survey data that informed the analysis in this report. This report was edited by Stacy Andell, a UN Online Volunteer mobilized through www.onlinevolunteering.org Understanding land dynamics and livelihoods in refugee hosting districts of Northern iii Uganda

Table of Contents

Introduction ...... 1

Land Access and Acquisition in Refugee-Hosting Communities ...... 2

Refugee Settlement and Physical Planning ...... 6

Land Use and Food Security ...... 7

Effect of Refugee Influx on the Environment ...... 11

Potential Threats to Uganda’s ‘Progressive’ Land Allocation Agenda ...... 14

Way Forward ...... 16

Appendices ...... 17 Understanding land dynamics and livelihoods in refugee hosting districts of Northern Uganda 1

Introduction

Establishing systems to protect and sustain the seekers from 0.15 million in December 2015 to livelihoods of refugees presents challenges at 1.4 million by January 2018iv; concerns on the the global, continental, regional and national optimal utilisation of allocated land to support levels. Given the changing landscape of food security and sustainable livelihood humanitarian response, host governments objectives; and management of community and development partners are beginning to expectations in return for hosting refugees. shift the philosophy of refugee assistance to 1 durable solutions for addressing humanitarian Focusing on Northern Uganda , this brief and development needs of refugees and their provides insights into land as the main host communities. In Uganda, this philosophy productive asset for building self-reliance. It started much earlier. The Government has, provides evidence on the access and utilisation since 2006, pursued a progressive approach of land considering the growing number of to refugee management backed by a refugees, also considering how the interplay conducive operational environment. First was of the local social, political and economic an inclusive Refugees Act of 2006 and Refugees context influences decisions on land. It also Regulations of 2010, which extended, among highlights the impact of land decisions on others, property rights, freedom of movement, household livelihoods and the environment the right to work, and access to services. and proposes options for optimising Uganda’s progressive refugee policy. Second, the National Development Plan (NDPII-2015/16-2019/20) articulates the Information was drawn from 12 focus group Refugee Settlement Transformative Agenda discussions (FGDs) and 16 key informant (STA) through which it seeks to foster sustainable interviews (KIIs) held in November 2017 among livelihoods through, in part, sustainable select host and refugee communities in management and efficient use of iland for , , and Lamwo districts. refugees and host communities. For 2016/17 The districts were purposively selected to reflect alone, Uganda’s annual conservative cost the age of settlements, refugee population as of protection, management and provision of well as the social cultural diversity (ethnicity) of essential services such as health, education, refugee hosting communities (see Appendix security, and environmental services for refugees 1 for the detailed methodology). The refugee both in-kind and financial was estimated at respondents included both recent arrivals US$ 323 millionii. Taking into consideration the (less than two years) and those in protracted 2 protracted nature of the refugee situation, situations (more than 5 years) . Where data enhancing productive capacities is envisaged permitted, comparisons were made between to build self-sufficiency given the unpredictable hosts and refugees. The qualitative information nature of humanitarian support. was complimented by data from the 2017 Food and Agriculture Organisation (FAO) survey, As part of the above inclusive reforms, the which was conducted in all the 5 refugee- Government of Uganda allocates land for hosting districts in Northern Uganda. Lastly, residential and agricultural purposes, to administrative information from the sampled refugees, irrespective of ethnicity and originiii. Local Governments (LGs) and the Office of While the Government’s intent is still on course, the Prime Minister (OPM) regional desk offices there is growing debate on the land question, was used to substantiate findings from the particularly on issues pertaining to: the rising aforementioned sources. uncertainty of land access owing to the growing number of refugees and asylum

1 Emphasis on the northern region is motivated by unique characteristics that it exhibits including: the highest refugee population (over 70%); grow- ing vulnerabilities resulting from the long period of armed conflict and susceptibility to natural disasters (UNDP 2015); high and persistent poverty partly compounded by the above factors; and increased occurrence of land conflicts following cessation of internal displacement (Refugee Law Project, 2012). 2 The cut off for protracted situations was guided by a 2003 UNHCR study. http://www.unhcr.org/research/working/3e2d66c34/solu- tions-sight-problem-protracted-refugee-situations-africa-jeff-crisp.html Understanding land dynamics and livelihoods in refugee hosting districts of Northern 2 Uganda Land Access and Acquisition in Refugee-Hosting Communities Majority of refugees access land through average of US$11 per acre each rainy season. Government This form of cooperative behaviour could be The allocation of land among refugees is harnessed by development actors to enhance guided by the Refugee Act (2006) and Refugee sustainability of livelihood interventions. In some Regulations (2010). Accordingly, most refugees cases, implementing partners have facilitated access land through OPM on behalf of the refugees and hosts to form joint farmer groups Government. The percentage of refugees to undertake agricultural projects on relatively that have acquired land through OPM per bigger pieces of land (minimum of 5 acres for a settlement area ranges from 80.5 percent in 25-member group). Adjumani to 94.3 percent in Lamwo (Table 1). Approximately 4.1 percent of refugees are Owing to the relatively high levels of integration reported to have accessed land through and peaceful coexistence, refugees have agreements with landowners, while close to 1 devised other innovative ways of accessing percent accessed land without agreements more land for production. One of these is through but with user rights (Table 1). For districts such bilateral agreements with host communities, in as Moyo, Adjumani and Yumbe, “just walked which case, land may be provided for free or in” seems to be an important channel through hosts may request modest rent in the form of which refugees access land. This accounted cash payments or share cropping. This is the for 7.6 percent of refugees, 4.2 percent, and case for refugees in more than 80 percent of 4.6 percent in the respective districts. This the settlement areas. For example, in Imvepi model is likely to cause conflict between host settlement, refugees access land at an communities and refugees if left unchecked.

Group farming in refugee host communities: Photo credit: FAO Understanding land dynamics and livelihoods in refugee hosting districts of Northern Uganda 3 Approximately 69,336 hectares of land has upon which a survey is undertaken by the been allocated for refugee settlement in government. Although no direct monetary Northern Uganda benefits accrue to land owners, communities provide land in anticipation of positive As of May 2017, 49,531 hectares had been reciprocity from Government or development allocated for refugee settlement in the region. partners working in the refugee space. These One year later, the land size has increased by expectations range from the sharing of social 40 percent. Overall, Yumbe and Arua have services and interventions that are provided allocated the biggest share of their community for refugees and hope of rewards in terms of land, accounting for 48 percent and 32 percent, employment or scholarship opportunities. To respectively (Table 2). About 75 percent of land some extent, communities that host refugees allocated to refugees in the region is used for share social services such as health facilities, settlement purposes while the remaining 25 schools and roads that are constructed in and percent is distributed between agriculture and around settlements. public services3. Hosts also receive livelihood support in the OPM acquires community land for refugee form of farm tools and they may also benefit settlement through Local Government from interventions targeted for integrated structures (comprising of both refugees and hosts) OPM works with the local leadership to identify livelihood groups. Sharing of resources is guided land for settling refugees. Landlords with big by the local integration strategy that abolished chunks of land at the Local Government the use of parallel services by refugees and the level are requested to offer part of their land 70:30 Government directive that entitles hosts to settle refugees. In some cases, land owners to a share of 30 percent of the interventions contact authorities with land offers. The land supporting refugees4. Although efforts are made acquisition process is concluded by signing to secure lives and livelihoods of both refugees a memorandum of understanding (MoU), and hosts, this is not without challenges5. Table 1: Modes of land access by refugees (%) Adjumani Arua Lamwo Moyo Yumbe Total Purchased 0.8 1.1 0.4 0.4 1.6 1.0 Inherited or received 3.8 1.6 0.4 2.0 1.1 1.7 Leased in 1.2 0.3 0.0 0.8 0.3 0.5 Just walked in 4.2 0.9 0.9 7.6 4.6 3.1 Do not know 0.0 0.0 0.0 0.0 0.3 0.1 Received from Government 80.5 91.3 94.3 84.5 86.5 88.1 Agreement with land use rights 6.5 3.3 1.8 4.0 5.4 4.1 Without agreement but with land use rights 1.2 0.8 1.8 0.8 0.3 0.9 Other 1.9 0.8 0.4 0.0 0.0 0.6 Total 100 100 100 100 100 100 Source: Computations based on FAO data (2017) Table 2: Land size per refugee settlement Settlement District Ownership May 2017 May 2018 Rhino Camp Arua Community 7,030 14,487 Imvempi Arua Community 5,659 7,897 Bidi Yumbe Community 25,420 33,402 Lobule Koboko Community 110 2890 Palorinya Moyo Community 7,800 5200 Adjumani Adjumani Community 3,312 3,760 Palabek Lamwo Community 200 1,700 Total 49,531 69,336 Source: OPM Head Office; Refugee Desk office, Adjumani and Arua (2018) 3 Refugee desk offices, Arua and Adjumani 4 The 70:30 rule excludes food assistance that is exclusively received by refugees. 5 For some interventions, it is not easy to apply the 70:30 formula, resulting in hosts not fully benefiting from this provision. The Understanding land dynamics and livelihoods in refugee hosting districts of Northern 4 Uganda The refugee influx has impacted land Out of the 6 settlement areas considered by allocations the study, representing more than 60 percent OPM allocates land at household level of the total refugee population in the region, 50 irrespective of household size. Although plot sizes percent received plots measuring 30*30, while vary by settlement, each refugee household, the rest received more. on average, receives a 30*30m plot. Prior to Gender differences in land access in the 2016, households were allocated separate refugee settlements arise from who is registered plots for residence and agricultural use. They as a household head in the OPM database. In were given agricultural land of about 50*50m households where a man, wife and children and a homestead plot of 20*30m in Arua while were granted refugee status at the same time, allocation in Adjumani was 30*30m for residential the men tend to be the decision makers on and 30*50m for agriculture6. With the surge in the allocated plots. In households where the numbers since 2016/2017, Government focus women arrived first without a spouse, plots are shifted to meeting residential needs (refer to allocated to them as registered heads and Table 3). For instance, except for the 30*30m they tend to remain in control even when their plot for settlement, no extra plots have been husbands join them. allocated for agriculture in Adjumani7. The allocation in There are high Refugees possess ‘only’ user rights on the Rhino and Imvepi is currently expectations by allocated land 50*50m for both courtyard hosting refugees While refugee households can exclusively use and small-scale backyard including the assigned plots, they do not possess power gardening while refugees in employment to own, sell, rent out or pledge the allocated Lobule and Bidibidi receive opportunities, land by Government as collateral for credit 30*30m for settlement social services from financial institutions8. There is no fixed and an additional 30*30m and other timeframe over which refugees can use the for farmingv. Overall, benefits allocated land. Their user rights are revoked households are allocated upon relocation, resettlement elsewhere Key Informant, an average of 30*30m- or upon return to their countries of origin. Yumbe. approximately 0.22 acres Upon refugees’ departure, landowners can (Table 3 and 4). repossess their land together with all non-public Table 3: Land allocation size (metres) before and after refugee influx of 2016 Before 2016 After 2016 Residential Agriculture Residential Agriculture Rhino camp 20*30m 50*50m 50*50m Invempi 20*30m 50*50m 50*50m Adjumani 30*50m 30*50m 30*30m Palabek Not applicable (NA) NA 30*30m Palorinya NA NA 30*30m Lobule NA NA 30*30m 30*30m Bidibidi NA NA 30*30m 30*30m Source: OPM Refugee Desk office Adjumani and Arua, 2018

Table 4: Average land size per household (acres) Adjumani 0.23 Arua 0.16 Lamwo 0.23 Moyo 0.13 Yumbe 0.23 Total 0.23 Source: Computation based on FAO data (2017)

increase in demand for social services because of the increase in population also compromises quality of particularly health services but also learning; dissatisfaction among hosts citing inequity with regard to accruing employment opportunities. 6 OPM desk offices – Arua and Adjumani 7 Adjumani Desks includes settlements in Moyo, Adjumani and Lamwo 8 These restrictions are enshrined in the Refugee Act of (2006) and Refugee Regulations of 2010. Understanding land dynamics and livelihoods in refugee hosting districts of Northern Uganda 5 infrastructure developments in place. However, social infrastructure (schools, hospitals, roads, there is contention regarding ownership of churches etc.) is permanently given away as woodlots planted on community land. It is a contribution by the community to the State. not clear whether these woodlots should be While no direct compensation is offered at taken over by Government or the landlords this point, the community continues to benefit upon repatriation of refugees. For example, from the public services offered in and around trees planted by refugees in Alere refugee settlements. In cases where refugees borrow or settlement in 1995 were handed over to the rent land from hosts through bilateral informal sub county in 2005 when refugees returned to arrangements, there may be a specified period their home countries. Land that is allocated for for land use (e.g. for a season).

Box 1. Show-casing the potential of investing in refugee hosting communities. The case of Mukwano Industries in

Farmers in Bidibidi, Yumbe district attend to their sun flower garden; Photo credit: DanChurch Aid During the 2017 Solidarity Summit for Refugees held in Kampala, the private sector was encouraged to identify refugee-hosting communities as business hotspots, rather than areas of low potential. The current engagement of Mukwano, showcased below, provides early evidence of this potential, and value of cultivating partnerships to sustain business as a win-win for both the private sector and local communities. DanChurch Aid (DCA), a Danish NGO, entered into a partnership with Mukwano Group of Companies to promote the production, processing and marketing of sun flowers in Yumbe District by implementing a two-year project funded by aBi Trust. The project, which commenced in January 2018, is targeting 6500 famers, 40% of whom are refugees. Mukwano factory, located in , has a capacity to process 200 metric tons of sun flower daily but is challenged by low production levels in the region. Under this partnership, the company is guaranteed an increased supply of input to sustain its operations, while farmers are assured of a ready market for their products at stable prices. Proceeds from the sale of the product are expected to enable farmers to address family needs but also invest in non-farm activities to smooth consumption during dry seasons. The factory provides training to farmers in sustainable production practices and quality control. To ensure quality, the seed is purchased directly from the factory, estimated at Shs 100,000 per acre. At the beginning of the first season, DCA registered 3700 farmers in 7 pilot sub counties, although only 200 participated in production, partly owing of uncertainties associated with the first season, as well as early stages of the partnership. For proceeding seasons, farmers in all the 12 sub counties are eligible to enrol into the programme. DCA has dedicated a gender and human rights specialist to address pertinent issues that may affect relations within households and among farmers. While refugee farmers access land from OPM particularly in Bidibidi Zones 1, 2 and 5; in some instances, DCA enters agreements with landlords for land on behalf of refugees for a period of 3-4 rainy seasons through agreements. According to the agreements, for every five acres provided by the landlord, DCA clears one acre for the landlord in return. In some locations, for instance, Zones 1 and 2 in Bidibidi, refugees have established partnerships with host communities to secure land for joint farming. DCA formalizes this process by facilitating simple time-bound agreements for land use to avoid future conflicts. The scheme shows the potential that exists for enhancing value chains and involving refugees and host communities at the various levels of the value chain. While the first production season is yet to come; the project demonstrated the opportunity to work with the private sector in livelihood diversification. This kind of scheme requires Government commitment to make it work. For example, one of the main challenges is the high cost of sunflower seed provided by Mukwano. Attempts by DCA to provide a 30% subsidy to vulnerable farmers have not been sustainable. It is imperative to devise mechanisms by which the factory provides seed on credit and ensures recovery during payment for farmer produce. Given that more than 70% of the host community are Muslims, with strong reservations regarding usury, credit should be provided interest-free. The Government should also consider providing additional extension services to enhance production and productivity to refugee and host farmers. Further effort to attract more private investors to refugee districts should be given due attention by devising appropriate incentive schemes. Understanding land dynamics and livelihoods in refugee hosting districts of Northern 6 Uganda

Refugee Settlement and Physical Planning

There is limited linkage between refugee wilfully ignored and hardly implemented. settlement and district physical planning Since planning is limited to town councils, Due to a lack of technical capacity and settlement of refugees is not guided by financial resources, physical planning in the district land use plans as none of the existing study districts has been largely limited to only settlements lie within the district town councils. town councils. Despite the existence of district Inadequate physical planning can pave the physical planning committees, planning is way for encroachment on protected areas undertaken in a piece- and wetlands or hamper urbanization efforts. meal and phased The district In addition, Local Government officials decried manner because of the limited involvement of relevant district physical planners limited financing. In and surveyors are departments in land mapping and refugee addition, the current settlement. not involved in the efforts in planning mapping of the have been largely Before settlement of refugees is done, UNHCR land for refugee funded by the donor undertakes surveys/land mapping to ascertain settlement in community through whether the secured land is suitable. These the district. OPM projects - which raises mappings involve determination of the land comes with its sustainability questions. size, holding capacity, soil characteristics, own surveyors and Individual interests social service sites, road networks and natural planners who are and political influence capital such as wetlands. Note that, this not locals to the were reported as other preliminary assessment is not always adequate. area factors undermining There have been reported cases of flooding in Adjumani Local physical planning in the Palorinya and Palabek refugee settlements in Government region; consequently, Moyo and Lamwo respectivelyvi, in some cases official. committee decisions are necessitating relocation of refugees. Understanding land dynamics and livelihoods in refugee hosting districts of Northern Uganda 7

Land Use and Food Security

While the quality of land allocated to refugees Reportedly, households in 83 percent of the varies by settlement based on availability, settlement areas studied fully utilise the plots most of the land allocated to refugees is allocated to them and require additional land being used effectively. to sustain livelihood. The rest account for a few cases such as Imvepi settlement camp where plots on rocky ground were not utilisable for crop Mindful of the fact that the number of refugees production. Overall about 30 percent received almost doubled between 2015 and 2017, the additional land outside of settlement camps for allocation of land is guided by availability cultivation. One such case included households and not land productivity. Fittingly, the quality in Bidibidi that were allocated land 6kms away of land varies across settlements. Some from the settlement. However, they have not settlements such as Mungula I & II in Adjumani, adequately utilised it, citing long distance, Palabek in Lamwo, are reported to have fertile limited access to inputs, and unsuitability of soils while others are reportedly rocky, infertile the land for agriculture. Understandably, this is and unproductive (e.g. Alere in Adjumani and pronounced for households headed or largely constituted by women and children. Although Ramogi, Kululu and Kochi sub counties in Bidi the distance falls within the 10 km standard Yumbe). Although no scientific soil fertility tests radius recommended by UNHCR, for labour- were conducted to ascertain the quality of soil deprived households, this distance poses a quality allocated to refugees, by and large, significant constraint to economic activity. In compared to hosts, refugees tend to cultivate the absence of official detailed mapping, it less productive land. was not possible to accurately quantify the extent of land use.

Table 5: Refugee livelihoods Settlement Size Use • Used for residence, back yard gardening and limited livestock rear 30*30m both residential land is fully utilised. Alere and agriculture • Households negotiate with host communities for additional land for and grazing.

30*30m both residential • Primarily used for residence, followed by backyard gardening (veg Mungula and agriculture fruits), and to tether a goat or two. The land is fully utilised.

• Residence and agriculture (50 percent allocation for each). Land is utilised. 30*30m –residential Bidibidi 30*30m – agriculture • Upon request, some households have received additional land for cultivation 6kms away from settlement but have not utilised it citin distance and inadequate farm inputs.

• Land in Palabek refugee settlement is primarily used for settlement 30*30m both residential Palabek kitchen and gardens take (10%) and 5% to tether a goat and some and agriculture Land is fully utilised.

• About 50% is allocated for each. Farming within these plots is mos backyard gardening and is maximally utilised. Rhino camp 50*50m both residential RS and agriculture • Households willing to access additional land are organised into farm groups which sometimes include local communities and provided l within not more than 2km from the settlements.

• About 50% is allocated for each. Households adequately utilise thi including farming, except for locations where the ground is rocky a 50*50m both residential Imvepi unfavourable. and agriculture • Given the space constraint, some households rent land from host communities at an average of US$ 10 per acre each season. Understanding land dynamics and livelihoods in refugee hosting districts of Northern 8 Uganda A high percentage of refugees identify Culturally, the different tribes specialize in themselves as farmers different economic activities. For instance, while the Dinkas prefer cattle rearing, which A majority of the refugees identify themselves requires significant land, the Madi and Kuku as farmers (Table 6). Among them, some are are predominantly inclined to crop production. farmers out of necessity rather than choice. Table 6: Refugee livelihoods Adjumani Arua Lamwo Moyo Yumbe Total Pastoralist 0.5 0.6 0 0 0.34 0.36 Agro-pastrolist 5.5 13.5 0.5 27 27.6 15.4 Farmer 94 70.4 99 67 71.4 77.6 Urban/Commerce 0 7.04 0 6 0.34 3.5 Others 0 8.4 0.5 0 0.34 3.3 Total 100 100 100 100 100 100 Source: Computation based on FAO data (2017)

Although refugees are more likely to use hosts, its implementation remains weak. This is improved seeds, technology uptake beyond evidenced by the low usage of improved seeds seeds is low; there is limited use of fertilizers among hosts. There is no effective mechanism and irrigation to monitor and ensure compliance with this policy, let alone foster accountability. On average, about 22 percent of refugees use improved seeds compared with just 8 Current land allocation cannot support percent among hosts. Use of improved seeds household food sufficiency in the absence is highest among refugees in Lamwo at 56 of food aid: percent followed by Moyo and Adjumani at 26 percent and 23 percent respectively (Figure Despite what Government and development 1). However, use of fertilisers and irrigation was partners offer, food sufficiency among refugees below 1 percent for host communities and remains just an aspiration. This can partly be refugees in all the five study districts. The high attributed to insufficiency of arable land for usage of improved seeds among refugees is refugees to supplement food production and attributed to interventions from implementing limited opportunities for alternative income partners. A high proportion of refugees access generating activities. This is exacerbated by the improved inputs as a package because of low productivity because of climate variability their refugee status. Despite the existence of a – unreliable rainfall, floods and droughts, policy that recommends a 70:30 ratio sharing but also the inadequate use of innovative of interventions between refugees and their modern technologies to make farming more Figure 1: Use of improved seeds and source (%)

Use of Improved Seeds Source of Seeds 90.0 90.0 80.0 80.0 70.0 70.0 60.0 60.0 50.0 50.0 40.0 40.0 30.0 30.0 20.0 20.0 10.0 10.0 00.0 00.0 Adjumani Arua Lamwo Moyo Yumbe Total Adjumani Arua Lamwo Moyo Yumbe Total Refugees Hosts Refugees Support Package Hosts Purchase/Rent Source: Computation based on FAO data (2017) Understanding land dynamics and livelihoods in refugee hosting districts of Northern Uganda 9 Table 7: Estimated production based on 0.22-acre (30*30m) agricultural plot a) Scenario 1a: Using traditional farming methods Finger Rice Beans Land (acres) 0.15 0.15 0.15 0.07 WFP ration person/month (kgs) 12 12 12 2.4 Yield (kgs/acre) 500 375 710 375 Production (kgs /year/household) 75 56.25 106.5 26.25 Estimated kgs/person/month 1.3 0.9 1.8 0.4 Gap (kgs) -10.8 -11.1 -10.2 -2 b) Scenario 1b: Using improved seeds Maize Finger Millet Rice Beans Land (acres) 0.2 0.2 0.2 0.1 WFP ration person/month (kgs) 12.0 12.0 12.0 2.4 Yield (kgs/acre) 875.0 500.0 710.0 460.0 Production (kgs /year/household) 131.3 75.0 106.5 32.2 Estimated kgs/person/month 1.82 1.04 1.48 0.45 Gap (kgs) -10.18 -10.96 -10.52 -1.95 Source: Authors’ computations based on 2017 crop productivity levels for Arua

sustainable, agricultural plot per refugee household. Two If one’s son marries and resilient, and cases are hypothesized – a 30*30m plot with and forms his own household, productive. without use of modern technology (improved he is not allocated a Consequently, inputs, such as fertiliser) and a 50*50m plot new piece of land, hence many refugees are with and without use of modern technology. the small piece of land still aid dependent The projections emphasise how the current allocated to the original irrespective of their low productivity levels amidst land constraints household is subdivided duration of stay. cannot sustainably meet the dietary needs to cater for the new Tables 7 and 8 of refugees in the absence of food rations. family. present scenarios The estimations are based on productivity on the estimated levels for and are computed for Refugee Welfare Council production cereals (maize, finger millet, rice) and beans Leader, Mungula RS. potential of (Table 7). Although it would be important to the allocated extend these scenarios to the other districts Table 8: Estimated production based on a 0.62 (50*50m) acre agricultural plot a) Scenario 2a: Using traditional farming methods Maize Finger Millet Rice Beans Land (acres) 0.43 0.43 0.43 0.19 WFP ration person/month (kgs) 12 12 12 2.4 Yield (kgs/acre) 500 375 710 375 Production (kgs /year/household) 215 161.25 305.3 71.25 Estimated kgs/person/month 2.99 2.24 4.24 0.99 Gap (kgs) -9.01 -9.76 -7.76 -1.41

Scenario 2b: Using improved seeds Maize Finger Millet Beans Land (acres) 0.43 0.43 0.19 WFP ration person/month (kgs) 12 12 2.4 Yield (kgs/acre) 875 500 460 Production (kgs /year/household) 376.25 215 87.4 Estimated kgs/person/month 5.23 2.99 1.21 Gap (kgs) -6.77 -7.76 -1.19 Source: Authors’ computations based on 2017 crop productivity levels for Arua Understanding land dynamics and livelihoods in refugee hosting districts of Northern 10 Uganda to capture agro-ecological and production would last a 6-member household about 1 and potential differences, comprehensive data 1.8 months, respectively. By implication, in the on productivity for different crops was only absence of food rations, a refugee household available for Arua district. The estimations can hardly sustain itself for a year by feeding on assume mono-cropping and a five-member food grown on the allocated plot. household. The choice of the crop combination was guided by the current WFP food ration that Although the food gap reduces with use of includes a cereal (maize or sorghum), beans, modern technologies9 and increase in land cooking oil, corn soya blend (CSB) and salt size, the quantities produced are still insufficient (see Appendix 2 for quantities provided per to meet food needs (Figure 8). category of refugee). We also assume that two thirds of the available plot would be allocated This has continued to happen amidst persistent to the cereal given the high dependence of aid reductions for refugees in protracted poor households on energy-dense foods and situations, owing to resource limitations. a third of the plot to production of a pulse Refugees whose food rations are reduced tend to adopt negative coping strategies such (beans). as reduction in number of meals per day and quality consumed (figure 2). Households that Estimations from Table 7 reveal interesting results. reported having spent an entire day without Assuming a refugee household allocates 0.15 a meal in the last 7 days preceding the survey acres and 0.07 acres of the allocated 0.22-acre ranged from 21.7 percent to 32.5 in Yumbe, agriculture plot to the production of maize and Arua and Adjumani against 2 percent in beans respectively (scenario 1a), holding other Lamwo and 6.5 percent in Moyo. This could factors such as weather constant, a household be explained by the fact that most refugees in would produce about 75 kgs of maize and 26 Adjumani and Arua are in protracted situations kgs of beans per year. Based on the current WFP and thus receiving less quantities of the food full food ration provided to each refugee (see rations depending on duration of stay. Appendix 2), quantities for beans and maize

Figure 2: Prevalence of food insecurity

Reduced number of meals in last seven days 9% Spent at least one entire day without eating in last 7 days (%) 32.5 75.1 74.1 73.5 29.58 67.1 62.3 59.5 23 21.72 42.0 39.0 37.4 16.67

7.79 16.2 6.5 2 0.62 0.6 Adjumani Arua Lamwo Moyo Yumbe Adjumani Arua Lamwo Moyo Yumbe Refugees Hosts Refugees Hosts Source: Computation based on FAO data (2017)

9 Refer to Appendix for projections assuming improved seed and fertilizer Understanding land dynamics and livelihoods in refugee hosting districts of Northern Uganda 11

Effect of Refugee Influx on the Environment

The current demand for natural resources is 90.7 million for 2016/17, constituting about 28 unsustainable. percent of the total public cost on refugee protection and management in Ugandavii. Due to a high refugee population density The contributing factors to ecosystem loss in settlements and surrounding areas, include land degradation, deforestation, loss environmental degradation has become a of vegetation cover, and water contamination matter of concern. The ecosystem loss due among others. to refugee settlement was estimated at USD Table 9: Refugee fuel wood needs per district Daily (tonnes) Annual (tonnes) Yumbe 1,007 367,666 Arua 889 324,618 Adjumani 850 310,259 Moyo 2,471 902,045 Lamwo 92 33,706 Total 5,310 1,938,293 Source: Computation based on FAO data (2017)

Deforestation is by far the fuelwood requirements per district based on Degradation of most observable effect the March 2018 refugee populations. the environment associated with refugees. has been taking Trees have been cut Such high demands on the environment place because down due to increased can easily lead to complete depletion of even mere fuel and construction aboveground biomass if left unchecked settling of the needs, land for without mitigating interventions such as refugees requires settlement and increased massive tree planting and use of energy mass clearing of commoditization of forest efficient cooking technology. For instance, the vegetation. resources by refugees in a hypothetical worst-case scenario, the Unfortunately, as a quick-short term estimated total existing above ground biomass this is income generating stock and growth from trees and shrubs in the unavoidable, activity. In Adjumani, Bidibidi settlement could supply fuelwood to Settlement about 75 percent of trees the current settlement population for up to only Commandant, have been cut for sale by three years (ibid). Arua farmers to provide poles and logs for construction To circumvent environment related challenges, during the refugee influx interventions and sensitization programs are in 2017, yet no more than 40 hectares of underway in many settlements and host trees have been planted to replace the cut communities. A common example is the trees. About 14,000 hectares of land have practice of tree marking, tree planting and the been degraded, yet only 900 hectares of the use of improved cooking technologies. Trees degraded land has been planted with trees in and around the refugee settlement have (ibid). been marked Red for “Do not cut” or Yellow for “can be cut only with special permission.” Almost all refugees rely on traditional biomass Red-marked trees are mainly fruit trees and for cooking and have limited access to trees with significance to the community in modern forms of energyviii To estimate the likely terms of size, land demarcation and religious effects of the high wood fuel consumption connotation. Yellow-marked and unpainted on the above ground biomass, the study trees may be cut and used for firewood. ICRAF uses Bidibidi’s average wood fuel daily and GIZ also promoted agroforestry for the dual consumption of about 3.5 kg per personix. role of food and environmental conservation; Table 9 shows the projected daily and annual however, in the short run, communities still Understanding land dynamics and livelihoods in refugee hosting districts of Northern 12 Uganda

Bidibidi before refugee settlement Bidibidi after refugee settlement Table 10: Annual fuel wood requirements; case of Bidibidi Mar-17 Mar-18 Mar-19 Bidibidi population 272,206 287,801 295,599* BAU Annual fuelwood consumption (Tons) 347,480 367,666 377,628 Annual charcoal equivalent (Tons)11 69,496 73,533 75,526

Annual fuel wood consumption (Tons) 243,236 257,366 264,339 Improved cook stove Annual charcoal equivalent (Tons) 48,647 51,473 52,868 *March 2019 is a refugee population estimate Source: Authors’ computations based on baseline data adopted from FAO (2017) Rapid wood fuel assessment for Bidibidi need to understand and appreciate the biomass in less than 3 years. importance of agroforestry. Transitioning to improved cook stoves12 with 30 Using a case study of Bidibidi refugee percent energy efficiency would reduce total settlement, the study evaluated the potential wood fuel consumption by 113,288 tonnes gains from using innovative energy saving per annum for Bidi bidi alone. This means that options at household level. Based on refugee even without tree planting, a move to the statistics of March 2017 and 2018, assuming a use of improved cook stoves would delay the 2.7 population growth rate10, we project the depletion of local biomass by a year. In terms refugee population of 295,599 by March 2019 of costs, with the business as usual scenario, the (Table 10). Using the 3.5 kg per person per day cooking fuel cost for a 6-member household biomass consumption (as business as usual was estimated at UGX 613,200 per annum while (BAU)), as at March 2017, the annual wood that for the improved cook stove would be fuel consumption for Bidibidi was estimated UGX 429,240. These estimates were computed at 347,480 tonnes. However, with population assuming constant prices of UGX 40,000 per growth projections the annual fuel demand 100kg bag of charcoal. A charcoal equivalent is likely to increase up to 377,628 tonnes per was adopted to estimate cost of because in annum threatening depletion of the total most cases firewood is obtained for ‘free’.

10 The 2.7% is a half of the population growth rate (5.7%) experienced between March 2017 and March 2018 11 Conversion to charcoal equivalents was used as a basis for costing following 672 tonnes of charcoal equivalent to 3360 tonnes of fuelwood FAO (2017) 12 Improved cook stoves refer to those that use charcoal, wood, animal dung or crop residue but in less amounts than tradi- tional cook stoves. In some cases, energy use efficiency is enhanced by clay lining Understanding land dynamics Understanding land dynamics and livelihoods in refugee and livelihoods in refugee hosting districts of Northern hosting districts of Northern 13 Uganda Uganda 13

Box 2: An Innovation with potential to save the environment: The case of ECO-STOVE Eco-stoves use volcanic rocks as an alternative for fuelwood in combination with and a small amount of charcoal which can be replaced by locally available materials like twigs, maize cobs, other crop residues and combustible home wastes excluding polythene. The stove is designed with an air system, which is run by solar. Installation involves an initial cost of US$ 244 for which a household acquires the stove, a solar panel, battery with an in-built radio, two 3-5watt LED lighting bulbs, a USB charger and 12kgs of volcanic rock stones reusable for up to 2years depending on household cooking practices. Whereas the use of local alternatives to charcoal dust has no direct cost implications for the household, we estimate that charcoal dust would cost the household up to US$ 24 per annum. This is based on a market rate of US$ 4 for a 100kg bag of charcoal dust which is an estimated requirement for two months. In the subsequent years, the household will have to buy only volcanic stones and charcoal with a contingency of US$ 14 for any arising maintenance requirements.

Although the initial cost appears high for an average refugee household to afford, this innovation becomes cheaper by the second year – costing a household only USD $ 57 for cooking fuel. For a population of 991,995 refugees in the region as of June 2018, the initial cost of this energy-saving investment is estimated at US$ 47 million, with anticipation of a drop to US $ 9.4 million in the second year. Moreover, additional benefits like income or cost saving from mobile phone charging, solar lighting, but most importantly, considerable reduction in wood fuel demand can be achieved, i.e. 0.6 tonnes of charcoal dust per annum compared to 75,526 tonnes and 52,868 tonnes of charcoal equivalents per household per year in the business as usual and the regular energy saving stove option. Understanding land dynamics and livelihoods in refugee hosting districts of Northern 14 Uganda Potential Threats to Uganda’s ‘Progressive’ Land Allocation Agenda

Unmet landlord expectations and protracted refugee settlement (Agojo refugee settlement) refugee situations in Adjumani was opened; the elders had refused By and large, Government depends on the to host refugees in their area and it took a lot good will of bona fide landowners to provide of negotiation and persuasion from high-level land for refugee settlement in Northern Uganda. Government officials to calm the situation (Key To protect asylum space and generosity of Informant Interview, LG). This landlords, there is an urgent need for Government is exacerbated by the protracted nature of to consider the landlords’ expectations earlier refugee situations. discussed. For instance, locals in the Palabek host community fear for the safety of their land, with Land sustainability concerns on land suspicion of potential grabbing by Government allocation officials when refugees leave. Some landlords Despite the pressure that the current refugee claimed exclusion from the land mapping population has exerted on existing resources, exercise, resulting in loss of the fertile portions Uganda is likely to continue receiving refugees in of the land. There has also been contention the coming years. The 2018 integrated refugee regarding land response plan projects that Uganda will receive gazetted as burial about 375,000 new refugee arrivals in 2018x. A time will come grounds, for which Of these, about 300,000 are anticipated to when people will no communities demand come from , 60,000 from Congo longer be interested in compensation and 15,000 from Burundi. Refugees from South giving land given the since it is regarded Sudan will likely be settled in Madi in Arua protracted nature of as unproductive. district and a further expansion of the Palabek refugee situations in Growing suspicions settlement in , while those from Uganda. Government have resulted in Congo and Burundi will be settled in Western should consider some communities Uganda (ibid). Subsequently, Government will permanently acquiring rescinding their have to secure more land in Northern Uganda to land in Northern decision to allocate accommodate the expected arrivals from South Uganda additional land Sudan. Table 11 shows that an extra 37,200 acres Commandant, to government. of land will be required to settle the projected Adjumani. A case in point is 300,000 refugee arrivals from South Sudan in 2018 where landlords had assuming a 50*50m allocation per household. provided land for agriculture to OPM Although still minimal, the prevalence of in Palabek settlement but later reclaimed it. land conflicts in refugee hosting districts is on To address these challenges, communities the rise expressed the need for Government to provide land titles to the owners as a safety guarantee. While conflicts are still minimal, ranging from 10 cases in Lamwo to about 20 in Adjumani in Since land is a fixed resource, it is highly unlikely 2017, they are generally on the rise in all districts that the current model of land allocation will (Figure 3). Conflicts between refugees and be sustainable in the long term. For instance, hosts exist but are not necessarily related to the Ciforo Sub county incident where the last land grabbing. Table 11: Projection of land size required to settle refugees in Northern Uganda Assumptions Land required Arrivals Based on projected refugee arrivals 300,000 No of households Assumes a 5-member household 60,000 Required land in acres (settlement only) If refugees receive only 30*30 plot (0.22 acres) 13,200 Required land in acres (settlement + agriculture) If refugees receive only 50*50 plot (0.62 acres) 37,200 Source: Authors computations Understanding land dynamics and livelihoods in refugee hosting districts of Northern Uganda 15 Only minor cases of trespassing were reported. compensation, after willingly offering over 2000 These typically arise when refugees cross their acres of land for refugee settlement. boundaries to cut grass to thatch their houses in host community lands, cut trees from the It is important to put the issue of land and neighbourhood for charcoal burning, graze on associated conflict in historical perspective hosts land without consent, source firewood as contentious issue such as land ownership without consent and enter into false leases– and land rights between the community, where some hosts pretend to lease land to Government and at times, private individuals refugees without consent from other family are context specific. Refugees hosting districts in members. Northern Uganda, following the displacement that took place due to past conflicts in the There are some reported cases of region have blurred ownership and created disagreement between OPM and landlords. access and utilization constraints in a number The disputes are largely triggered by individuals of cases. This situation should be taken fully into in pursuit of personal benefits as opposed to account as it creates some new dynamics in community benefits (Commandant, Bididbidi). the refugee response management. For example, there is a pending court case over a health centre Normally there that was constructed Figure 3: Prevalence of Conflict by OPM are conflicts over in Bangatuti village, 20 whether the land was Zone 4 Odravu Sub allocated for refugee County, Bidibidi. The landlord who signed 16 settlement or not, 13 whether the land the general offer 11 lord was consulted, document offered 10 10 whether social service the land in question to OPM; but the 7 structures and 6 permanent structure transaction is disputed by a new claimant should be built and 3 over the same land. In wrong perception and 000 000 beliefs of land lord another instance, the Araba clan members Adjumani Arua Moyo Lamwo Yumbe (KII, Arua) have approached the president seeking 2015 2016 2017 Understanding land dynamics and livelihoods in refugee hosting districts of Northern 16 Uganda

Way Forward

(i) Graduation from food aid and/or reduction refugee relief efforts, but also enable benefit in food aid should be considered together from sustainable livelihoods solutions. with promotion of improved agricultural (vi) To avert further destruction to the environment methods and provision of additional and mitigate the likely negative effects, there arable land: While the generous refugee is a need to promote consistent, continuous policy provides for the allocation of land and sustainable land use and management for settlement and agriculture in refugee practices from the onset. Ongoing efforts hosting districts, some of the factors thwarting such as ordinances against bush burning sustainability of refugee livelihoods include: and the sale of charcoal in the small size of agricultural plots; disregard can be scaled up. For sustained protection for family size in land allocation; low quality of the environment, restoration should be of land in some settlements, low adoption of embedded and mainstreamed in LG work technologies and heavy reliance on rainfall. plans and budgets. (ii) To permit for optimum use of land, it is (vii) As a matter of urgency, development imperative to profile, map and zone land actors must promote affordable renewable in refugee hosting areas so that promotion cooking technologies to replace the use of production interventions is based on of fuelwood and charcoal. In so doing, the peculiarities of the different areas. By there is a need to ensure that the quality ensuring that agricultural enterprises are of the technologies introduced is certified. matched with climatic and soil conditions, Transitioning to cleaner and more modern the different production zones will benefit energy solutions will reduce fuel costs at the from the comparative advances that household level and save the environment Uganda’s agro-ecological zones offer. Land that both refugees and host communities Profiling would facilitate identification of depend upon. specific interventions required to enhance production and productivity. (viii) While physical planning at the LG level is provided for in the 2010 National Physical (iii) There is a need to promote and expand Planning Act, it remains one of the unfunded interventions to support alternative priorities. Financing is crucial for ensuring livelihood strategies for both refugees and organized settlements but also paving hosts to sustain livelihoods, rather than focus the way for organized urbanization or the on dependence on agriculture, which is creation of satellite towns. Funding should land intensive. Putting in place effective be complemented by capacity building to incentives to attract private sector actors is enable LGs to sustain the land use planning. key in diversifying livelihoods. The institutional linkages between LGs, OPM (iv) Development actors should facilitate and UNHCR in the implementation of land mechanisms for group formation among use plans must be strengthened. hosts and refugees. Mechanisms that (ix) The expectations of host communities and will guarantee security of land rights and land owners must be managed to avoid production and proceed management will conflicts. Doing so requires, in part, ensuring not only sustain social cohesion but also fair that the land offers are adequately and participation. professionally negotiated and hosts fully (v) To promote social cohesion, it is imperative understand the terms. Government officials for Government and development actors to must also avoid making unreasonable deliberately plan for and include hosts in their pledges, based on which the community programming. Besides having expectations, can hold Government accountable. host communities, in many cases, face similar Government should consider settling vulnerabilities as refugees. Implementation refugees on public land to avoid of the of the 70:30 ration in refugee response negotiation encumbrances. Introducing interventions is pertinent for incentivising host land administration and certification communities to continue contributing to the programmes might hold significant benefit for both hosts and refugees. Understanding land dynamics and livelihoods in refugee hosting districts of Northern Uganda 17 Appendices

Appendix 1: Data and Methods The study employed different research methods. A critical review of relevant literature was undertaken to contextualize land ownership, access, use and livelihoods in Northern Uganda. Qualitative data was collected through key informant interviews with a cross section of resource persons and focus group discussions with refugee and host communities. This was supplemented by data from a survey conducted by FAO (2017). This study was carried out in 4 refugee-hosting districts including: Adjumani, Arua, Yumbe and Lamwo. The districts were purposively selected to reflect age and size of the settlements as well as the social cultural diversity (ethnicity) of refugee hosting communities. For example, although the settlement in Lamwo is relatively new and small, it has a population that is culturally different from the West districts and has a different political history and landscape which could impact on land use and livelihoods. On the other hand, Adjumani and Arua are large settlements with populations of both protracted and recent refugees, which is critical for assessment of self-sustenance as proposed in the NDP II’s refugee STA. In Adjumani, Ayele and Mungula I were sampled; these are relatively old settlements with protracted and recent arrivals and, a reasonable percentage has been able to access land for agricultural use. In Arua, Invempi and Rhino camp were sampled, Rhino is a relatively old settlement while Invempi is relatively new. In Yumbe and Lamwo, we visited Bidibidi and Palabek refugee settlements, the only settlements in these districts. Communities within a 15km radius of the settlement were considered hosts; the choice of the 15km radius cut off was adopted from a 2017 study by UNHCR13.

Data collection summary Instrument type Study participants Purpose Understand land ownership, access, land use practices 12 FGDs in total (3 per district) • Protracted refugees (more than 2 years) and livelihoods • Recent refugees (less than 2years) • 4 with host communities • Women and men To identify linkages if any between land and livelihood • 8 in refugee communities • Opinion leaders outcomes for the different communities and groups in each community • Settlement commandants To provide expert insights on key issues related to • Refugee leaders livelihoods and sustainable land use 30 Key Informant Interviews • District leaders • Cultural leaders To identify opportunities and fragilities with current • Refugee desk officers livelihoods and land use approaches

Appendix 2: Monthly WFP food rations and cash grants to refugees Food Rations Item Time of arrival into Uganda 2015-2017 2012-2014 2009-2011 Cereal (maize or sorghum)/Kgs 12 6 3 Cooking oil/Kgs 0.9 0.45 0.45 Beans/Kgs 2.4 1.2 0.6 Corn Soya Blend (CSB)/Kgs 1.5 0.75 0.75 Cash Grant Option (UGX) 31,000 17,000 Extremely vulnerable 45,000 irrespective of date of length of stay Source: Interview with refugee welfare council leaders, 2017

Appendix 3: Production projections based on improved seed and fertiliser 0.23-acre plot with improved seed and fertilizer Maize Finger Millet Rice Land (acres) 0.15 0.15 0.15 WFP ration person/month (kgs) 12 12 12 Yield (kgs/acre) 1875 920 1420 Production (kgs /year/household) 281.3 138 213 Estimated kgs/person/month 3.91 1.92 2.96 Gap (kgs) -8.09 -10.08 -9.04

13 UNHCR (2017) Livelihoods socio-economic assessment in the refugee hosting districts Understanding land dynamics and livelihoods in refugee hosting districts of Northern 18 Uganda

0.62-acre plot with improved seed and fertilizer Maize Finger Millet Rice Land (acres) 0.43 0.43 0.43 WFP ration person/month (kgs) 12 12 12 Yield (kgs/acre) 1875 920 1420 Production (kgs /year/household) 806.25 395.6 610.6 Estimated kgs/person/month 11.2 5.5 8.5 Gap (kgs) -0.8 -6.5 -3.5

Appendix 4: Profiles of refugee settlements in the study districts West Nile, Arua District –Rhino camp refugee settlement Rhino Camp is in Arua district and covers an area of approximately 225 square kilometres; it is divided into 10 zones and has 42 villages (or clusters). Rhino camp is one of the oldest refugee camps and was established in the 1980s. However, due to the influx of refugees in 2016 and 2017, Rhino Camp settlement was extended to cater for new arrivals and is currently about 75 square kms. Before the refugee influx, each refugee household used to be allocated a residential plot of 30*30m and cultivation land of 50* 50m. Currently OPM is allocating 50*50m for both settlement and backyard garden.

West Nile, Arua District – Imvepi refugee settlement Imvepi settlement in Arua is a relatively new settlement. It was opened in February 2017 after Palorinya settlement in Moyo district, which was opened in December 2016, had reached its refugee-hosting capacity. Both Imvepi and Rhino Camp settlements currently host 223,097 refugees, comprising 21 percent of the Arua district population. Imvepi previously hosted South Sudanese refugees and was considered a suitable location for the new settlement as some of the previous infrastructure was still intact and landlords were willing, once again, to host refugees. Like in other settlements, OPM provides refugees with land for settlement.

West Nile, Adjumani District – Adjumani refugee settlements Adjumani district has had about 18 refugee settlements established in different years. Some are relatively small and old while others are big. The settlements include: Ayilo 1 and 2 established in 2015, Mungula 1 & 2 (1996), Alere 1 & 2 (1990), Olua 1 & 2 (2012), Pagirinya (2016), Baratuku (1991), Baroli (2014), Nyumanzi (2014), Maaji 1, 2 & 3 (1997), Ayilo (2015), Mireyi (1994), Oliji (1991) and Elema (1992). The current refugee population stands at 239,335, constituting 59 percent of Adjumani’s population. Some households have been allocated a plot of land for settlement and agriculture while others just have plots for settlement.

West Nile, Yumbe District – Bidibidi refugee settlement Bidibidi is in Yumbe district near the South Sudanese border. The settlement opened on 3rd August 2016 to help decongest Adjumani, which had been overwhelmed by an influx of refugees from South Sudan. Between 3rd August and December 2016, Bidibidi received about 272,168 refugees, making it the biggest refugee settlement in Uganda. Currently, Bidibidi is home to about 285,014 refugees, comprising about 34 percent of Yumbe’s district population. The settlement is divided into five zones spread across the five sub counties of Ramogi, Koki, Kululu, Odravu and Ariwa. The majority of households in Bidibidi have so far been allocated a settlement plot of 30 *30 metres and are yet to receive additional plots for agricultural use.

North, Lamwo district – Palabek Refugee Settlement Palabek refugee settlement in Lamwo district was officially opened in April 2017 to enhance reception capacity of new arrivals. It is one of the newest refugee settlements in Uganda and currently hosts about 35,533 refugees constituting 20 percent of Lamwo’s population. The settlement is clustered (with host community integrated with refugees) and is estimated to cover about 51 square kms. Palabek settlement is in the three sub counties of Palabek Kal, Palabek Ugili and Palabek Gem and is so far, the only refugee settlement in the Acholi region. Understanding land dynamics and livelihoods in refugee hosting districts of Northern Uganda 19 Endnotes

i Office of the Prime Minister, GoU (2015), “A concept paper for the Settlement Transformative Project 2016-2021 ii UNDP (2017), Summary of Study: Uganda’s contribution to refugee protection and management. iii http://www.judiciary.go.ug/files/downloads/Act%20No.%2021of%202006%20Refugees%20Act2006.pdf iv https://ugandarefugees.org/category/policy-and-management/refugee-statistics/?r=48 vi Government of Uganda, 2011 Physical Planning Guidelines. vii UNHCR (2017) Uganda Operational Update on the South Sudan Refugee Situation, Inter-Agency Update 1st-7th June 2017 viii UNDP (2017), Summary of Study: Uganda’s contribution to refugee protection and management. ix Lahn, G. (2015). Heat Light and Power for Refugees: Saving Lives, Reducing Costs. Chatham House. x FAO and UNHCR, (2017). Rapid wood fuel assessment: baseline for the Bidibidi settlement, Uganda. Rome, Food and Agriculture Organization of the United Nations (FAO) and Geneva, Switzerland, United Nations High Commissioner for Refugees (UNHCR). xi Office of the Prime Minister, UNHCR (2017), “Uganda Refugee Response for 2018: The integrated response plan for refugees from South Sudan, Burundi and the Democratic Republic of the Congo Understanding land dynamics and livelihoods in refugee hosting districts of Northern 22 Uganda

United Nations Development Programme (UNDP)

Plot 11, Yusuf Lule Road, Find us on; Nakasero P.O. Box 7184 www.ug.undp.org Kampala, Uganda. facebook.com/undpuganda Tel: +256 417 112100/301 twitter.cowmww/undpuganda Fax: +256 414344801 Email: [email protected]