Livelihoods in crisis

Livelihoods in crisis: a longitudinal study in Pader, Uganda Ellen Martin, Celia Petty and James Acidri HPG Working Paper October 2009 About the authors Ellen Martin is a Research Officer with the Humanitarian Policy Group at ODI. Celia Petty is co-founder and Director of Operations of Evidence for Development. James Acidri is a livelihoods development specialist with the livelihoods development group (LDG) in Uganda. Acknowledgements This three year study owes its completion primarily to the patience and good will of the participating households, whom we gratefully thank. We would also like to thank the Pader based staff of Mercy Corps, and in particular Lawrence Wao Acellam, David Amone and Roger Horton who have been involved with the work since 2006 as well as Ammar Zakri; and to Zoe Daniels and Darius Radcliffe, country directors of Mercy Corps at various stages of the study. Our thanks also extend to the wide range of individuals and organisations who gave their time and assistance to the researchers. Finally, we would like to thank Susanne Jaspars, Paul Harvey, Sara Pantuliano, and Liz Stites for their comments and help in drafting. John Seaman’s support in data analysis and charting and his observations over the three year study have been invaluable, as were Kevin Savage’s insights and enthusiasm in launching the project. About the Humanitarian Policy Group The Humanitarian Policy Group at ODI is one of the world's leading teams of independent researchers and information professionals working on humanitarian issues. It is dedicated to improving humanitarian policy and practice through a combination of high-quality analysis, dialogue and debate. Humanitarian Policy Group Overseas Development Institute 111 Westminster Bridge Road London, SE1 7JD United Kingdom Tel: +44(0) 20 7922 0300 Fax: +44(0) 20 7922 0399 Website: www.odi.org.uk/hpg Email: [email protected] ISBN: 978-0-85003-918-4 © Overseas Development Institute, 2009 Readers are encouraged to quote or reproduce materials from this publication but, as copyright holders, ODI requests due acknowledgement and a copy of the publication. This and other HPG Reports are available from www.odi.org.uk/hpg 1 Contents Acronyms ........................................................................................................................................................3 Introduction ........................................................................................................................................5 1. 1.1 Overview ...................................................................................................................................................5 1.2 Methods ................................................................................................................................................6 2. The conflict and its humanitarian impact.............................................................................................7 3. Humanitarian relief and recovery.........................................................................................................9 4. Livelihoods in transition in Pader ......................................................................................................11 4.1 The households studied .................................................................................................................11 4.2 The return process ..........................................................................................................................11 4.2.1 Patterns of return ....................................................................................................................11 4.2.2 Decision-making about return .................................................................................................12 4.3 Security, return and livelihoods ......................................................................................................12 4.4 Changes in livelihood strategies .....................................................................................................13 4.4.1 Livelihoods before and during displacement ...........................................................................13 4.4.2 An overview of livelihood strategies after 2006 .......................................................................14 4.4.3 Analysis of household economy in the study population .........................................................14 5. Access to assistance and services.....................................................................................................23 5.1 Access to basic services .................................................................................................................23 5.1.1 Government capacity...............................................................................................................23 5.1.2 Access to services for the study population.............................................................................23 5.2 Humanitarian and development assistance ....................................................................................24 5.2.1 Agricultural assistance ............................................................................................................24 5.2.2 Livestock support....................................................................................................................24 5.3 Credit and Savings..........................................................................................................................24 6. Conclusions ......................................................................................................................................27 Appendix.......................................................................................................................................................29 References ....................................................................................................................................................31 List of illustrations and boxes Chart 1. Increase in household food production, 2006–09 (data for the 23 households assessed in all three study years ........................................................15 Chart 2. Crops sold by surveyed households, 2008–09................................................................................17 Chart 3. Sources of food income, 2006–09 ..................................................................................................19 Chart 4. Changes in Disposable Income/Adult Equivalent 2006-09 ..............................................................20 Box 1. Adela and David ................................................................................................................................16 Box 2. Claire and Christopher.......................................................................................................................18 Box 3. Lillian ................................................................................................................................................21 2 Acronyms CAP CMR DRC EVI GoU HPG IDP IHM LRA NRM NUSAF OCHA PRDP SACCO TRS UGX UNPRAP UPDF WFP WHO Consolidated Appeal Process crude mortality rate Democratic Republic of the Congo extremely vulnerable individuals Government of Uganda Humanitarian Policy Group internally displaced person individual household model Lord’s Resistance Army National Resistance Movement Northern Uganda Social Action Fund Office for the Coordination of Humanitarian Affairs (UN) Peace Recovery and Development Plan Savings and Credit Co-operative transitional resettlement site Ugandan shilling UN Peace Building and Recovery Assistance Plan for Northern Uganda Uganda People’s Defence Force World Food Programme World Health Organization 3 1. Introduction highlighting people’s vulnerability to changes in external assistance, as well as to other factors, including inflation and climatic conditions. The report detailing the findings of the second year of this study (Savage et al, 2008) called on the humanitarian community to recognise the diversity of needs of different households as well as the fragility of the recovery process. It was also concluded that continued humanitarian assistance would be required even following a peace agreement and people’s return home, as many households take some years to rebuild productive livelihoods. 1.1 Overview This working paper is the third report in a threeyear study that has monitored changes in the livelihoods of a small group of conflict-affected households in northern Uganda since the beginnings of a peace process in 2006. As people move on from camp-based livelihoods and return to their home villages, the paper looks at the continuing challenges they face in resettlement, at how the population has been supported in their recovery, at what has been effective and what more could be done. This last round of monitoring in January and April 2009 focused on the implications of changing security, assistance and governance environments for livelihoods and recovery. The report also provides insight into factors that facilitate household recovery, from the perspective of the capacities of the households themselves as well as from that of the Government of Uganda (GoU), humanitarian and development donors and agencies. It examines the challenges households face in meeting basic needs, including food security, and raising sufficient income to invest in livestock, education and farm inputs as the aid architecture shifts from one of humanitarian assistance towards recovery and development. Little is known about livelihoods transition and transformation in post-conflict contexts. Most research and analysis in emergencies has been conducted on a single-study, snapshot basis, and affected populations have rarely been followed over time. To address this gap, the Humanitarian Policy Group (HPG)—in partnership with Evidence for Development and Mercy Corps—initiated a study that would follow a small group of households in Pader district with the aim of conducting in-depth qualitative and quantitative research in order to investigate the main economic, social and political issues affecting their livelihoods. While it focuses on two small areas and is not designed to describe specific needs throughout the region, its findings are relevant to the evolving policy and programming context in the region. Following the introduction, the report provides a brief overview of the conflict in Chapter 2, focusing on elements that are most relevant to people’s livelihoods in Pader. Chapter 3 is concerned with the policies of the GoU, international donors and aid agencies that are shaping the livelihood recovery process. Chapter 4 presents an in-depth analysis of the changes in livelihood strategies of a small group of households following the cessation of hostilities in 2006, and the continuing and emerging sources of instability. It looks at their household economies, productive capacity and income levels, as well as land use and livestock holdings. Chapter 4 also analyses the key factors that have influenced decision making about return, the characteristics of households that have fared best in the recovery process and those that have faced the greatest difficulties. Chapter 5 analyses the assistance provided by the GoU and different humanitarian and development agencies, both national and international. The final chapter draws together the conclusions about key factors in recovery and The key questions for the research included: 1. How have people’s livelihood strategies changed over time? 2. How does the governance environment affect people’s livelihoods, particularly in relation to security and the availability of and access to assistance and services. 3. How significant is relief as a part of people’s livelihoods? 4. What are the key factors in the success or failure of household transition from a context of displacement to return? Findings from previous years showed that despite the improved security and increase in food production since 2006, many people were scarcely managing to access minimum food and non-food needs. Indeed, some households had experienced a reduction in their standard of living since 2006, 5 attendance. Household economy information was supported by detailed biographical/life history information. In addition to monitoring the household economy of the 23 households this last round of research also carried out focus group interviews and key informant interviews, including with youths, community leaders, local council officials, camp authorities and security forces (police officers and military units). Interviews focused on expectations for recovery and changes experienced over the past three years regarding security, access to resources (including assistance, land, livestock and employment) and how these influenced livelihood strategies. Key informant interviews were also carried out in Pader town centre and in Kampala with representatives from district government, UN agencies, national and international NGOs as well as donors, focusing on the governance environment and the institutional transition from humanitarian relief to recovery and development assistance. offers recommendations on the type and level of assistance needed for people living in Pader and those living in similar circumstances in other parts of the country. 1.2 Methods The study initially followed 28 households from two communities in Lira-Palwo sub-county, Pader district, Uganda1 (23 households were studied in 2008–09 as contact was lost with five). These households cover a range of wealth groups and are illustrative of the diversity of vulnerabilities as well as capacities of people to promote their own recovery. The study used the individual household model (IHM) to measure the economic status of the sample households on an annual basis. The Appendix provides details of IHM methodology.2 Households in the sample were visited every six months to update information on employment and production, changes in demography and school 1 For details of selection methodology see Petty et al (2007), pp. 13–14. 2 See also, www.evidencefordevelopment.com. 6 2. The conflict and its humanitarian impact The conflict in northern Uganda has been ongoing since the mid-1980s (Bradbury, 1999; Lucima et al., 2002; Finnstrom 2008; Dolan et al., 2006; International Alert, 2008). There have been 22 armed rebellions in Uganda under President Yoweri Museveni (Ruaudel and Timpson, 2005), and the emergence of the Lord’s Resistance Army (LRA) can be seen against the background of the formation of several insurgent groups as Museveni’s National Resistance Movement (NRM) took control of northern Uganda in 1986. By 1990, the LRA was the only significant insurgency remaining in the region and it quickly lost any support among the population, which was the main target of LRA attacks (Allen, 2005). which camps were to receive assistance (Branch, 2008). Few humanitarian agencies were operating in northern Uganda until 2004. Many people were therefore not receiving any assistance at all (Dolan et al., 2006). With overcrowded conditions, poor sanitation and little access to water, mortality rates in the camps were significantly in excess of the emergency threshold for crude mortality of 1/10,000/day. A study carried out by MSF Holland in Pader and Lira in 2004 found the overall crude mortality rate (CMR) to be 2.79/10,000/day and under-five mortality rates at 5.40/10,000/day (MSF Holland, 2004). A study led by the World Health Organization (WHO) in 2005 across Acholiland found the CMR to be 1.54/10,000/day and under-five mortality rates at 3.18/10,000/day (RoU, 2005). The GoU followed a counter-insurgency strategy of forcibly displacing the population. The aim was to isolate the LRA from any potential sources of support and to make it an easier target for UPDF operations throughout the region (HRW, 2005). Officially people were placed in ‘protected villages’ set up to shield the population from LRA attacks. In practice this strategy had a disastrous impact on people because they were separated from their livelihoods and received limited assistance. By 2004, Pader district was home to 24 camps hosting 290,000 people, or around 90% of the population (MSF Holland, 2004). In 2005 an estimated 1.5 million people across Acholiland, or 80% of the population, were living in camps (Allen, 2005). The humanitarian crisis finally caught the attention of the international community. In 2003, then Under-Secretary-General for Humanitarian Affairs Jan Egeland had made his much-publicised statements about northern Uganda as one of the worst humanitarian crises in the world; two years later, the UN Security Council also put pressure on the Ugandan government to address the situation. For the Ugandan government, election losses and the prospect of the International Criminal Court turning its attention towards the human rights abuses it had committed were additional incentives for a change in strategy. In August 2006, the LRA and the government of Uganda signed a ceasefire, which was followed by peace negotiations in Juba (Sudan) between 2006 and 2008. Since 2006 there have been no reported LRA attacks in northern Uganda. Forced displacement was associated with looting of property and destruction of livestock, both by the LRA and the Uganda People’s Defence Force (UPDF) (Dolan, 2000). Within the camps, livelihoods options were restricted because movement outside of the camp security zones meant being at risk of attacks by the LRA and being targeted as a rebel by the UPDF. The camps themselves were a source of insecurity as the army and Local Defence Units failed to provide adequate protection from LRA attacks and committed many human rights abuses (UN, 2008). The 2006 ceasefire agreement, and improvements in security from 2005, allowed people to leave the camps after years of displacement to start rebuilding their lives and livelihoods. By July 2009 it was estimated that more than 850 000 people— almost 80% of the population in Acholiland—had left the camps and resettled in transit sites or in their areas of origin. In Pader, just over 11% of the population is believed to still reside in IDP camps (UNOCHA, 2009b). In the absence of government assistance, humanitarian agencies started providing relief (mainly food aid) to internally displaced persons (IDPs) and thus became drawn into the encampment policy. Distributions were accompanied by military convoys, and a selective approach was adopted by the government as to Despite these positive signs, security is still a major issue for a number of reasons. At a local level, this is due to an increased incidence of banditry and growing conflicts over land as people return home. Return in the absence of adequate 7 assistance and services has led to hikes in malnutrition and mortality rates and increased levels of family separation in some contexts (UNOCHA, 2007a). appear to be underrepresented, which may reflect the observed pattern of child mortality (Medair Uganda, 2007). At the national level, the ceasefire is unravelling, and this is leading to increased regional instability. By November 2008, LRA leader Joseph Kony had failed to appear at three separate attempts to sign the final peace agreement. The Juba peace talks officially ended in December 2008, an event that was marked by the launch of Operation Lightning Thunder against LRA camps in Garamba Park in the Democratic Republic of the Congo (DRC) by the Government of Uganda in partnership with the DRC and Southern Sudan, as well as with support from the United States Africa Command (AFRICOM). Population demographics from mortality surveys show an under-representation of men and women between the ages of 20 and 40 (RoU, 2005; Medair Uganda, 2007). Many people were abducted and killed in the war. The conditions in the camps are also an important factor, the absence of young women in the 20-25 age group for example can be partly attributed to very high rates of maternal mortality . Others—particularly young men and women—have left for urban centres and in some cases districts outside the conflict zone. Children under the age of five also 8 3. Humanitarian relief and recovery The peace process and relative security in northern Uganda has led to significant changes in the aid environment, particularly a move towards recovery and development after years of emergency relief assistance. Humanitarian funding to Uganda markedly increased from 2006 as international attention to the humanitarian crisis in Uganda grew and security improved. In 2003, for example, contributions amounted to $135 million, which increased to $279m and $326m in 2006 and 2007, respectively, and to $316m in 2008 (UNOCHA, n.d.). example that the PRDP understated peace and reconciliation in favour of technical solutions. Specifically, it only allocated a nominal 2.7% of the budget to its fourth objective, and emphasised peace building and reconciliation at the micro-level over national reconciliation (RLP, 2008). There is also the question of whether the PRDP actually represents additional funding or if it is just a ribbon to put around existing allocations. In the case of Pader district, for example, regular central government transfers for 2008/09 amounted to UGX 20 billion (approximately $11.4m), yet for 2009–10 this sum has decreased to UGX 16.3bn ($9.3m) and PRDP allocations of UGX 3.9bn ($2.2m)3 effectively serve just to fill the gap (RoU, 2009). Similar trends have been noted for other districts (RLP, 2008). Expecting a peace agreement to be signed, donors started pushing for a transition from relief to recovery and development. UNDP’s office of the Resident and Humanitarian Coordinator in Uganda developed an Early Recovery Strategic Framework in 2007 with the aim of laying the foundations for a successful implementation of the governmentled Peace Recovery and Development Plan (PRDP). There has been little visible progress since the PRDP’s official launch in 2007 and UNDP’s early recovery strategy also gained little support with donors, who expressed doubts over UNDP’s capacity to lead the process. Humanitarian agencies began shifting their approaches towards recovery in 2006–07, but the scale of returns and basic service and infrastructure needs in return areas were far greater than humanitarian agencies had the capacity to meet. Until recently, the PRDP had no funding mechanism. Given the experiences of previous recovery programmes, such as the Northern Uganda Social Action Fund (NUSAF) I, which suffered from massive corruption, there has been much discussion among donors about how to track financing. This is of particular concern for direct budget support donors. The UK’s Department for International Development (DFID), for example, currently channels 50% of its assistance through direct budget support. In the absence of central government efforts, two trends have emerged on the part of development donors: either direct support to local government and international agencies by donors such as the US Agency for International Development (USAID), which opened offices in the north, or, as is the case for many donors, a continuing preference for bilateral support, which does not necessarily reach intended beneficiaries in the north. The PRDP is the outcome of efforts to put in place a government-led post-war recovery plan through four strategic objectives: the consolidation of state authority; rebuilding and empowering communities; revitalisation of the northern economy; and peace-building and reconciliation. Under the three-year framework, the government is to contribute 30% and donors 70% towards a total budget of just over $600m; three-quarters of that budget will be allocated to districts through central government and donor funding mechanisms (RoU, 2007). In the absence of clear UN leadership on recovery and no dedicated funding mechanism, the 2008 Consolidated Appeal Process (CAP) had sought to integrate both early recovery and recovery activities under the ‘Parish Approach’, a strategy between the GoU and the UN Office for the Coordination of Humanitarian Affairs (OCHA) to provide assistance on the basis of geographical area, including increasing access to basic services, supporting livelihoods and rule of law There has been widespread confusion as to the procedure for the operationalisation of the framework, little consultation with stakeholders at the district and local levels and a lack of prioritisation of activities (RLP, 2008). Official implementation did not start until July 2008, when the government pledged 30% of the $606m budget. More important concerns have been raised, for 3 Figures are based on average yearly exchange rates provided by www.oanda.com. 9 and developing infrastructure and civil administration (UNOCHA, 2007a). Although some humanitarian donors saw this as a positive step, it was also perceived as “an inappropriate extension of the humanitarian mandate at best, and institutional preservation at worst” (Bailey, Pavanello, O’Callaghan and Elhawary, forthcoming). Financial support was limited, and at t the mid-year review, most of those projects were therefore removed from the CAP and have had little impact on the ground. At the same time there has been little interest in creating a separate mechanism specifically oriented towards recovery. Influenced by the strong development donor relationships with the government, relief is likely to be phased out rapidly and replaced by development funding, which has been slow to materialise. Building and Recovery Assistance Plan for Northern Uganda (UNPRAP), aimed at supporting the PRDP through human rights, justice reconciliation, local governance, social services and livelihoods and social protection programmes (UNCT Uganda, 2009). All sectors place a strong emphasis on supporting a government-led recovery process. The year 2009 has therefore been a turning point, marking a clear move towards a nationally led recovery process. The 2009 CAP has a very narrow definition of humanitarian aid and funding sought for its projects has been reduced by 40% compared to 2008, from $374m to $247m, of which $101m has been requested for the Acholi region. Yet by the mid-year review in June, only 45% of total requirements had been met, and only 41% in Acholiland (UNOCHA, 2009a). Insufficient funding has been a major challenge for agencies, with some considering closing operations. This narrow approach to humanitarian programming is reflected in a focus on food aid distribution to extremely vulnerable individuals (EVIs), including the elderly, orphans, female-headed households and people with disabilities. Even this targeted approach has been difficult because of limited funding. While 766,000 EVIs were to receive food rations, only a 100,000 have received them so far in 2009. The new country programme of the World Food Programme (WFP) will entail recovery plans aimed at supporting the government’s priority areas for agricultural growth and the strengthening and diversification of livelihoods (WFP, 2009). The PRDP strategy was re-launched on 1 July 2009, with greater support from the GoU, including almost $500m towards monitoring and implementation mechanisms. The PRDP will mainly be implemented through central and district government structures; however, concerns remain about its focus on large-scale infrastructure and economic development. A number of development donors are planning to support activities in the north, but it remains unclear whether there will be a comprehensive donor approach. The World Bank has launched NUSAF 2, which is focused on the economic revival of the north. Following district-level consultations, the UN Country Team has devised a new UN Peace 10 4. Livelihoods in transition in Pader This section describes the return experiences of a small group of households studied over the past three years and the main challenges they have faced in re-establishing their homesteads and livelihoods, particularly from 2008 to 2009. It focuses on the changes in livelihood strategies and the impact of the current political, security and aid environment on their livelihoods. from January 2007, when a primary school opened. Lira Palwo camp remains the main local market hub on which Obolokome TRS depends for trade, social services and employment opportunities. By the time of our last assessment in April 2009, all households from the original study group had left the Lira Palwo camp and the majority were back in their home villages.4 One of the villages under review, Aringo Pee, is located within the boundaries of Obolokome TRS, about 4km from the main IDP camp in Lira Palwo; the other, Aywee Keyo, lies around 8km from Lira Palwo. Members of the study group from Aringo Pee had all resettled, either on their original land, or within Obolokome TRS. For many, remaining in Obolokome TRS is the preferred option as they are already close to their land and can cut the cost of further relocation by staying put. Most of the study group originally from Aywee Keyo have also returned to their home village, although the process has been slower for them. Three out of the 11 Aywee Keyo households in the study moved back in 2007–08, followed by five in 2008–09. Only one household moved directly from Lira Palwo to Aywee Keyo while the other returning households went first to the TRS. One elderly household head originating from Aywee Keyo has been resettled by her extended family close to Lira Palwo camp and is happy to remain there with their support. Two other households from Aywee Keyo are still living in Obolokome TRS. Both households are headed by women, and although they would like to return, they are unable to do so due to a lack of labour to construct huts. 4.1 The households studied The households are members of two villages in neighbouring parishes in Lira-Palwo sub-county, Aringo Pee and Aywee Keyo. From the original study group of 28 households, eight households fled to Lira Palwo in 2001. Lira Palwo had not yet been designated as an official camp but was perceived to be safer as it was a trading centre and less isolated. The majority of households—15— were forced to move following the start of mass displacement and the government encampment policy in 2002 following Operation Iron Fist, a military offensive against the LRA. They were followed by five households in 2003 and remained displaced until peace talks and the resettlement process began in 2006. 4.2 The return process 4.2.1 Patterns of return As people started to leave the camps across the sub-county, a total of 209 transit sites were created. According to OCHA population monitoring data from July 2009, it is estimated that of a total population of almost 23,000 in Lira Palwo, around 30% reside in transit sites and only 15%, or just under 3,000 people, still remain in the older IDP camps (UNOCHA, 2009b). The majority of those who remain in the Lira Palwo camp are members of the host community or lack the necessary labour and cannot afford the costs involved in relocating. Several of the households have children who have migrated to urban centres within the district or the region, as well as to Kampala. The households reported very limited communication with them since they left. Research by the Refugee Law Project estimates that there are between 300,000 and 600,000 ‘urban IDPs’ in Acholiland (RLP, 2007). During 2006, the first households started to leave the camp for the transitional resettlement site (TRS) in Obolokome, which had been established in January 2006 as part of overall plans to decongest and eventually resettle IDPs. NGO projects, including the provision of seeds and tools, cash for work and water and sanitation projects started in Obolokome at this time. The pace of resettlement to Obolokome TRS increased 4 Three households from the original study group have moved to other locations. Of these, two have remained in the subcounty (one individual remarried). The third household head returned to her original home in another part of Pader district. 11 also the sense that some people living in Obolokome TRS felt that it was now time for those households not originally from the parish to leave. In addition, messages from government officials signalling that households should return were often perceived as orders rather than an administrative communication. 4.2.2 Decision-making about return Decisions around the timing of return for households in the study group, as for the wider population, were determined by four key factors that varied in importance according to individual needs and priorities: the opening of schools; the availability of thatching grass for roofing; the availability of labour—mainly during the dry season (December to March)—to build houses; and a sense of security, which was particularly relevant for return to the more remote Aywee Keyo. The importance of education has been a recurrent theme in this study; many families waited until the primary school had reopened in their community. Not all households have made a complete break with the camps. As is a trend in the rest of the district, the distance between a household’s original home and the location of the camp is often just 5km to 10km (UNOCHA, 2007b). Some households commute on a daily basis between their homes in the TRS and their land, or between the TRS and Lira Palwo, which is also a trading centre with a secondary school and a health facility. For those living in Aywee Keyo, distance has a greater bearing upon access to services. The distance to Lira Palwo is 8km and the track is completely impassable throughout the rainy season. This restricts access to both the nearest health centre, which is in Obolokome TRS, and to secondary education in Lira Palwo. The only option for students who want to complete their education is to attend as boarders, thus adding to the cost. Some households that had planned to move back to Aywee Keyo in the second year of the study (2007–08) were set back because fire destroyed much of the thatching grass they had acquired and they lacked the means to replace it. Building new homes is expensive and, for the elderly in particular, the timing of return was dependent on family members being able to help with the construction of huts, as they could not afford to hire labour. The availability of labour was an important factor for households more broadly as they needed to clear land as well as construct huts and, at the same time, earn as much income as possible from casual labour. Finally, households moving on from the TRS to Aywee Keyo wanted to complete the agricultural cycle on land rented or borrowed in Obolokome. A range of factors therefore determines how and when individual households move to their home areas. While access to services plays a part, it is by no means the only consideration. For example, this year some households chose to return to Aywee Keyo when the primary school opened, despite the lack of clean water and although they could access clean water in the TRS. Similarly, food aid has been blamed for slowing down the return process. This study, however, found no evidence that the availability of food aid acted as a disincentive to return; the majority of households needed and were entitled to receive food rations regardless of whether they had relocated. While ongoing security concerns did not prevent households from returning, they played an important role in determining how and when they moved. As noted in the last report, some of the first households to return to Aywee Keyo settled around the school building and borehole rather than in their more remote homesteads. Some members of households who were interviewed individually said that while they would have liked to move earlier, they waited until enough of their neighbours shared their confidence in security and moved back in a group. Nevertheless, many remained worried by the risk of unexploded ordnance, particularly after three children were killed in a neighbouring sub-county at the end of 2008. 4.3 Security, return and livelihoods A sense of security was one factor in decisionmaking on return. Perceived risks to safety included retaliatory attacks by the LRA, banditry, stock theft and conflict over land. Land disputes are a serious problem, and across the region there have been a growing number of reports of violent fighting between clans, of huts being burned down and of incidents of mob justice. Regarding push factors, there were reports over the past two years of pressure from host communities on some households to return to their own land. This was mostly mentioned in relation to Lira Palwo, and is confirmed by reports on return monitoring (ASB, 2008). Yet there was 12 4.4 Changes in livelihood strategies In Lira Palwo as elsewhere, the government spoke out on the radio after the launch of Operation Lightning Thunder in December 2008, announcing that there was no risk of the LRA returning to Northern Uganda and that Joseph Kony and other top LRA commanders would be apprehended. Messages of reassurance from government concerning Operation Lightning Thunder were viewed with scepticism as they had also been heard after the earlier Operation Iron Fist. As one discussion reflected: 4.4.1 Livelihoods before and during displacement Before displacement, members of the study population were predominantly subsistence farmers. The main cereals grown were sorghum, maize, finger millet and rice; root crops, including cassava and sweet potatoes; oil crops, including soya, groundnuts, sunflower seeds and simsim (sesame); and pulses such as cow peas and green beans. The main cash crops were cotton and oil seeds. Around half of the study households cultivated cotton; all households cultivated some oil seeds for sale. Some households also collected shea nuts to make butter, which they sold locally. Fishing and hunting with dogs were important activities, and a number of households listed fishing equipment among the assets they had lost. government promises to fight Joseph Kony and then to sign a peace agreement and has again changed its policy. We hear that they are fighting, the government says it’s no problem but the radio [broadcast] is not enough. The absence of a peace agreement as well as the risk that the army might fail to apprehend the LRA leader was a source of concern. All households in the study group owned some livestock, which included cattle and plough oxen, goats, pigs and poultry. Non-agricultural occupations among the study population included one carpenter, three masons/bricklayers and two teachers, all of whom combined off-farm employment with their own agricultural production before they were displaced in 2002. In eight households (around 30% of the sample) women earned income from brewing. The local economy was vibrant due to good access to trade and traditional markets such as Lira Palwo, Okwang and Patongo as well as regional market hubs such as Lira Municipality, Kitgum and Gulu. Cash crops such as cotton were marketed through local cooperatives such as the Lira Palwo Cooperative. During 2008, armed banditry was often a primary security concern, and one that has grown over the past two years. Referred to as boo kec (Acholi for ‘thugs’), bandits were also varyingly described as demobilised soldiers, ex-LRA fighters, or people disguised as LRA fighters to cause confusion. Households reported on incidents ranging from theft of food to cattle raids and armed ambushes. The government has sought to address these concerns by increasing the number of civilian police posts in the region by recruiting Special Police Constables. Many of these recruits are exmembers of the local defence units that were created during the conflict; others are former LRA fighters. Yet salaries are minimal and frequently delayed, and resources for investing in security are lacking. There are also concerns over corruption within the security forces and possible involvement in cattle thefts. Nonetheless, the investment in Special Police Constables was highlighted as a positive step by several households; two household members reported that a son and a brother-in-law were employed in the police force. For many young men, the military and police provide one of the few available livelihoods strategies. During displacement, assets were lost or destroyed, nearly all the livestock were killed or stolen and people were unable to access their agricultural land to grow food and cash crops. Few people were able to access credit and remittances were almost unheard of, other than as one-off gifts from visiting relatives. Many IDP households engaged in petty trading, particularly of firewood and charcoal during the dry season, and wild greens and seasonal fruits during rainy seasons (Stites, Mazurana and Carlson, 2006). Other livelihood options included brewing and stonequarrying (FAO, 2006). These limited economic activities were concentrated in the main IDP camps such as Lira Palwo and Omot (Acidri, 2007). 13 for poor timing of seed distributions, poor quality of seeds and a shortage of hand tools. Nevertheless, 2007–08 saw a significant increase in crop production, and flooding only affected the area after most of the main food crops had been harvested. 4.4.2 An overview of livelihood strategies after 2006 The transition away from camp-based livelihoods began in 2006, when people started to leave the camps. Land access was extremely limited in the Lira Palwo IDP camp. Households could only produce their own crops by renting or borrowing land around the camp from the host community. High rental costs meant that poorer households cultivated tiny pockets of land with short-term maturing crops, and even for the better-off, cultivated plots were small. Shortage of land for cultivation and lack of employment meant that living standards were extremely low; at the time of the first assessment, the four poorest households surveyed could not meet minimum household food requirements,5 with around 63% falling below a local standard of living threshold (Petty et al, 2007). WFP food aid—consisting of maize, beans and oil—initially provided a basic ration covering 60% of food needs. This was reduced to 40% during the second year of the study (starting in April 2008) and general ration distributions stopped in December 2008, with food aid continuing only for EVIs. From the start of the resettlement processes in 2006, WFP distributions were available both to the remaining IDP camp population as well as to those who had returned to their villages. A mid-year re-enumeration exercise in 2007 meant that a number of households had their food aid entitlement cut, causing a drop in their ‘food income’ and living standards. This came at an inopportune time, as most households were extremely busy during this second year of resettlement, with house construction and the clearing and cultivating of their own land. These activities left very little time to earn cash from employment. Households received little practical assistance when they moved from the main camp to Obolokome TRS. Early NGO interventions included the opening of a borehole by Mercy Corps in the second half of 2006, a cash-for-work road-clearing project and a tree nursery and grain storage projects6 in Obolokome TRS. There was limited support from the GoU in the form of agricultural tools. 4.4.3 Analysis of household economy in the study population The pace of movement out of the main IDP camp increased sharply in January 2007 with the opening of a primary school in the TRS. Assistance towards livelihood rehabilitation in the following months included NGO seed fairs and seed distribution projects.7 Participants in the study and other community members criticised the project Income through food production. Food production increased sharply in 2007–08, the first year in which crops were harvested from the newly accessible land outside the immediate camp area in Lira Palwo; further increases followed in 2008– 09 (see Chart 1). 5 The WHO reference standard for the minimum daily food requirement is 2,100kcal/person. 6 By 2008 the tree nursery had been abandoned. Respondents explained that it had not been possible to maintain the project as it was not a priority for them, and there were other demands on their time. 7 For details, see Petty et al (2007), Acidri (2007) and Savage et al (2008) 14 Chart 1. Increase in household food production, 2006–09 (data for the 23 households assessed in all three study years) In the most recent survey year (2008–09), despite a 35% increase in land area cultivated, production (based on the kcal value of food produced) only rose by around 20% compared to the previous year. This is partly explained by low levels of rainfall. Poor returns also point to a lack of animal fertiliser and difficulties clearing and cultivating with hand tools. There were no programmes directly supporting land clearance and tillage in the study area until 2008–09, when Mercy Corps set up a voucher-based project that targeted the very poor (see below). Over the past three years, the area cultivated by households has increased from an average of 1.3 acres (1/2 hectare) in 2006–07 to 3.1 acres (1.25 hectares) in 2007–08 and 4.2 acres (1.7 hectares) in 2008–09.8 However, according to individual household interviews, the majority of households are not yet cultivating all the land that they own or to which they have rights. For most households in this survey, therefore, availability of land for cultivation is not in itself an issue; rather, the problem is one of land clearance, due to the lack of draft power and labour. 8 15 Each ‘study year’ runs from 1 April to 31 March. Income from employment. The change in income from employment has been less marked than the increase in household ‘food income’ from production over the study period. Average employment income per year dropped from UGX 133,186 in 2006/7 to UGX 70,697 in 2007–08 when most households were heavily involved in land clearance and house construction; however, income from employment rose again in 2008–09 to UGX 150,518, mainly due to sale of some crop harvest, construction work and better access to natural resources such as charcoal, thatching grass, local saving schemes and fishing. There is no indication that employment income will increase during 2010 (and there could be an overall reduction with the departure of many NGOs from the district). The main types of work available are likely to remain unchanged: agricultural labour, sale of grass and firewood (done by poorer women), construction work, including brickmaking (better paid), and fetching water (low pay, mostly female labour). Demand for skilled artisans is low and both the skills and opportunities for higherpaid salaried work are in short supply. However, there is scope to raise household income from agricultural production through the sale of cotton and other cash crops, and, over time, through livestock sales. Box 1. Adela and David9 This household has progressed steadily over the study period. This young couple had just married when they were displaced from their parents’ land in 2002. Their three children were all born in IDP camps—the last two after they moved to Obolokome. They were given three acres (1.2 hectares) to cultivate in Obolokome and moved to land allocated to them in Aywee Keyo in December 2008, where they now have access to 4 acres (1.6 hectares). This household also benefits from employment generated by NGOs: they receive regular help from a brother working for an international agency distributing food aid. They earn money from casual labour and brewing, as well as from some crop sales. The main reasons for their relative prosperity are the presence of two able-bodied and economically active adults, their smaller-than-average household size and their access to enough land to produce food for their own consumption as well as some cash crops. Adela is enterprising and hopes to open a kiosk but lacks the savings to do so. The household’s main concerns are the lack of draft power, which they would use to clear more land, and the lack of seeds to increase their production and cash income. Adela does have security concerns, although these did not prevent them from moving away from the TRS to a more remote location. Her worries include: 1. Ongoing fighting between the GoU and the LRA. 2. Rampant cases of robbery and theft. 3. The lack of clean drinking water in Alwee Keyo, which has forced them to drink the dirty river water. This is major concern for the household. 9 Income through sale of crops. In 2007–08 only five households in the study group sold crops, with a total value of UGX 31,700. In 2008–09 the number of households increased to 16, with a total value of UGX 2,338,160, mainly attributable to cotton sales (see Chart 2). Names have been changed throughout. 16 Chart 2. Crops sold by surveyed households, 2008–09 The number of households with chickens has fallen from just under half in 2007–08 to none at the last count, although poultry numbers vary widely through the year due to endemic poultry disease. These returns are still low and reflect a) poor climatic conditions in 2008–09, since a lack of rainfall earlier in the season resulted in complete failure for some crops; b) shortage of draft power for land clearance and tillage; and c) lack of cash to invest in cotton inputs. The returns highlight the continuing need for capital investment, notably in plough oxen, and for multi-year programmes to prime the agricultural economy. The weather is a key factor in the rate of recovery and, in these conditions, the premise that ‘self-sufficiency’ can be restored to agricultural communities within an 18-month aid cycle is simply erroneous. The World Bank-funded NUSAF restocking project, implemented across Pader district during 2008, was intended to boost tillage and land clearance capacity. Yet the project was widely criticised by intended beneficiaries as few people were able to access support. Only two households in the study group acquired cattle through the NUSAF scheme. Pig numbers have dropped, but there has been an increase in the number of goats owned by the survey group. As households begin to restock, the lack of veterinary services is a major concern, with at least two households reporting the death of valuable plough oxen in the past year, in addition to the loss of poultry and pigs due to disease. Income through livestock. Livestock, and particularly cattle and oxen, are essential for the re-establishment of a viable and potentially prosperous agricultural economy in this region (Petty et al, 2007). Before they were displaced, all households in the study owned some livestock and 50% of households owned cattle, with numbers ranging from two to 50 head. Only one household reported any surviving animals (a single cow) from the pre-war years. There has been some increase in the number of livestock owned by members of the survey group in 2008–09, but these cases are extremely limited, mainly due to the inability to afford to buy cattle rather than concerns over potential cattle raids (see Table 1). Land clearance and accessing draft power remains a top priority for households in the study group and for the wider population, yet many are still held back by difficulties in acquiring livestock, tools and draft power as well as cash to employ labour. Household members cited all these problems when they were asked about their investment priorities. Many referred to the ‘pitiful’ gesture of a handful of hoes received from the GoU in 2007. These were of such poor quality and so few in number that the community distributed them by lottery. Table 1. Survey group livestock holdings, 2007–09 Cattle/calf Poultry Pigs Goats 2007–08 7 43 7 1 2008–09 10 0 5 7 17 The problem of land clearance as a precondition for resettlement has not been sufficiently addressed by humanitarian agencies. Farmers have been left to do the best they can, opening up land for cultivation mostly with hand tools; in most cases, the land is not cleared in time to bridge the gap created when food aid is withdrawn. Only now, in the third year of ‘peace’, are ox ploughs gradually appearing in the richest households. The role of food aid. The period 2006–09 saw a shift from an overwhelming reliance on WFP food aid in 2006 to the current situation in which households’ own production has increased significantly. Yet food aid continues to play a major role in meeting household food needs as income from employment and crop sales remains low. In the first year of study, well over 70% of ‘food income’ was derived from WFP food aid. By the third year of study, food aid accounted for around 40% of ‘food income’, reflecting the reduction in food rations and increase in own crop production (see Chart 3). Low levels of income from the sale of last year’s crops and other employment has also meant that there is a shortage of cash to hire labour, to buy fertiliser and pesticides (essential inputs for cotton) and to purchase livestock. Box 2. Claire and Christopher This household of a wife, husband and four children has been doing well, but the recovery gains remain fragile. Before the conflict, they cultivated 10 acres (4 hectares) and owned 50 cattle. All their livestock was lost in Karamojong raids. In the first assessment, this household was in the lower half of income distribution. They cultivated 1 acre (0.4 hectares) in the IDP camp at a rent of UGX 20,000 and grew sweet potatoes, simsim, sorghum and groundnuts to supplement WFP rations. The household transferred to Obolokome in January 2007, when the primary school opened, and cultivated 4 acres (1.6 hectares) of land, which they were able to borrow without payment. In February 2008 they moved back to their original homestead, which is in a remote location, three miles from Aywee Keyo. Their priority, on returning to their own land, has been to increase agricultural production and raise cash income from groundnuts, simsim and cotton. This strategy has so far been successful. In 2008, they cultivated 8 acres (3.2 hectares) and sold more cotton than any other household in the study group, which can be partly explained by the high fertility of their homestead land and by their access to draft power. Nevertheless, fortunes have been mixed. One of the oxen they purchased for UGX 350,000 died from cattle disease in June and, although the husband continues to receive additional income as an NGO health volunteer, if the project closes, this income will be lost. The household head is a carpenter; if there is local demand, he could potentially increase the household’s cash income through work in this trade. The household currently has three children in primary school and one child of preschool age. When questioned about her main concerns, Claire did not cite security as a problem, despite their remote location. Her main worries were: 1. Lack of safe water in the area of resettlement. 2. How to replace the oxen that died. 3. Lack of health care services in the village of resettlement. Factors contributing to the household’s success include demography—two able-bodied and hard-working adults who both contribute to household income; the capacity to earn additional income from artisan and other skills (carpentry, volunteering at an NGO); and a culture of saving in the household (both husband and wife are members of saving schemes). A ‘sober’ lifestyle may also help: the wife gave up brewing when her husband became a Christian in the IDP camp. The loss of oxen worth UGX 350,000 is a very serious setback and underlines the need for improved access to veterinary services (as well as improved policing of cattle raiders). School fees will add to household expenses if the children go on to secondary school. Due to the remote location of the homestead, the children are likely to board at secondary school, adding significantly to the cost. 18 Chart 3. Sources of food income, 2006–09 for agricultural rehabilitation or investment. Many poorer households will simply not purchase more expensive items such as cooking oil and may also have to reduce expenditure on basic non-food needs. Food aid has been critical to resettlement; without it, people could not have survived for the first two years while rebuilding their homes and livelihoods. However, for most of the study population, WFP food aid distributions stopped in December 2008, nine months ahead of the date that had originally been announced.10 Just two of the surveyed population received EVI rations in January, and one elderly household head was told she would start receiving rations in April. In cash terms, the average cash value of food received during 2008–09 was around UGX 157,175, at local market prices. Recalling the income gap highlighted above, the main challenge for most households in 2009 will be replacing— with their own production or purchased food— around 300kg of WFP maize rations, which had provided an average of around 40% of household energy needs in 2008. It is unclear how this will be achieved given the limited local employment opportunities and extremely low levels of support Differences in recovery within the study households. Despite the constraints experienced by people returning home, there was a slight increase in the standard of living between 2006 and 2008. Chart 4 illustrates the changes in disposable income over the three-year study period. Ranking (1–23) is based on disposable income in the first year of the study (2006–07).11 Overall, the analysis shows that, while disposable incomes have dropped for a few households (mostly due to changes in food aid entitlements after a WFP re-enumeration in 2007), the majority of households are somewhat better off now than they were in 2006, when they were still confined to the main IDP camp and could not access their own land. 10 11 For methodological notes, see the Appendix of this paper and Petty et al (2007). This premature discontinuation was due to WFP funding and pipeline constraints. 19 Chart 4. Changes in Disposable Income/Adult Equivalent 2006-09 Some households have been able to recover more quickly than others. A number of factors appear to influence successful recovery, including the availability of labour in the household and, to some extent, assets and skills acquired before displacement. Characteristics of the poorest and richest households are shown in Table 2. assessment, when four of the ten households at the top of the income distribution had either a single or disabled household head, now only one of the top ten households falls into this labour poor category; the household in question is headed by a disabled woman. With more ablebodied labour (both male and female), households can cultivate their own land and at the same time maximise cash income from other sources; consequently, they are able to invest in livestock, rent ox ploughs for cultivation and invest in inputs for cotton production. before the war; the richest was a teacher and the next richest had around 50 head of cattle. As more land becomes available for cultivation, there is evidence that the economic differentiation between labour-rich and labour-poor households is increasing. In contrast to the 2007–08 Table 2. Characteristics of poorest and richest households Poorest quartile More likely to be headed by single or disabled person Income from sale of grass and firewood No cattle Do not sell crops, do not brew The poorest households all have low labour capacity relative to household size, with three out of the poorest six households headed by a single, elderly, or disabled person. Three households in the poorest quartile (2008–09 assessment) owned no cattle or small stock such as goats before they were displaced, and do not appear to have had skilled or semi-skilled off-farm income. By contrast, the very poorest household in the study held 40 head of cattle before the war; they sold cotton, the head of the household was a skilled mason, and his wife brewed. However, the husband became disabled during displacement and, with eight dependents and only one ablebodied worker, their situation is extremely fragile. Richest quartile Married couple, around 30 years of age Income includes salaried or skilled work Own cattle Sell crops Better-off households, with incomes between UGX 100,000 and UGX 300,000, are now selling cash crops and earning additional income from skilled and semi-skilled employment. In the richest quartile (2008–09 assessment), the wealthiest household is headed by a teacher; all households in this group are headed by married couples in their late 20s and 30s. The two richest households had assets or skills that suggest both were well off An additional challenge that many households face is the obligation to look after orphans. More than half of all households in the sample are 20 It is clear from focus group discussions that households caring for orphans do consider this to be an ‘added burden’. In some cases additional costs such as school fees are paid by other relatives; anecdotal evidence suggests, that without this assistance orphans may have less chance of completing education than other children in the household. However, the critical problem in access to education is not orphan status as such but affordability. Household income remains the most important factor in determining access to both primary and secondary education; it was cited by parents across the income range as a major concern. caring for orphans. Table 3 shows that the greatest number of double orphans (children who have lost both parents) are found in Q3, towards the top of the distribution. Single orphans (children who have lost one parent) are spread fairly evenly across the distribution. Q2, the next poorest quartile, has the largest number of households caring for single orphans. Table 3. Orphans in the household Q1 (poorest) Households with single orphans Households with double orphans Total households Q2 Q3 Q4 (richest) 2 3 2 1 1 0 3 0 3 3 5 1 Box 3. Lillian Lillian has been among the very poorest households in all the assessments. She was displaced from Aringo Pee to Lira Palwo with her husband and children in 2002. Her husband died in the camp in 2003 and since then she has cohabited and had another child. Her older daughter has also had a child, whose father is unknown. Even before displacement, this was a poor household. They did not own any livestock and lived off three acres of land (1.2 hectares), producing various food and cash crops such as sorghum, simsim and cotton. Lillian also earned some additional income from brewing but borrowed the equipment as she never had funds to buy her own. They were unable to grow any food in Lira Palwo due to the high cost of renting land and were wholly dependent on food aid. The household shifted to Obolokome in January 2007 and cultivated two acres (0.8 hectares) of land. Lillian was unable to benefit from NGO cash-for-work programmes in 2006 and 2007 as she could not take time out from her own cultivation. Although they missed out on the 2007 seed distributions, the household was included in the 2008 NGO tillage support programme. She used the tillage voucher (worth UGX 63,000), to cultivate simsim, a relatively highvalue oil crop. However, yields were very low due to poor weather, and she harvested just 60kg, which has a total value of around UGX 60,000—less than the cost of cultivation. The household has retained 5kg of seeds for 2010 and plans to keep the rest for its own consumption. Lillian also received a seed voucher worth UGX 40,000, which she used to purchase groundnuts. This crop also failed, due to dry weather. When asked about her priorities, Lillian, like members of many other households, says she would spend any additional income to open up new land, and to pay for education. Currently, none of her three children currently in primary school has a uniform and her younger daughter dropped out after just one year in primary school. She is very concerned that, as the other children get older, she will not be able to keep them in school. For Lillian, the most positive recent change has been to access her own land ‘without any hindrance’. However, she has not yet rebuilt a home there, due to lack of labour. Her main worries relate to securing labour for construction of huts in the home village and a lack of money for school fees as the children grow up and get to higher classes of learning. Lillian’s main problem is the lack of labour power. Although she gains some cash income from selling firewood, brewing and doing agricultural work, this is not enough to support a family of seven. Although she was included in the most recent seeds and tillage project, poor weather conditions meant that assistance with a cash value of more than UGX 100,000 yielded a cash return of zero. 21 5. Access to assistance and services 5.1 Access to basic services their quality of life and ability to meet basic needs. In the first year of this study, more than half of the households under review did not have sufficient income to meet a minimum set of non-food needs; as a result, many had to make trade-offs between purchasing basic items such as salt, kerosene and clothing, spending on social services—education or health care—or investing in seeds or basic hand tools to clear their land.12 5.1.1 Government capacity Local government in Pader district faces huge challenges in fulfilling its responsibilities for service delivery and promoting socio-economic development throughout the district, raising concerns over the reliance the above approaches place on national channels. Pader district was only created in 2001, and local government had little opportunity to strengthen its administrative structures before it was replaced by systems of camp and military leadership. In some cases disbursements from the central government have been delayed and the district leadership has pointed to a general decrease in funds both from the central government and international donors and agencies over the past few years. The cost of health care varies with individual household needs. Outpatient health services, including consultations in Obolokome and Lira Palwo health centres, are generally free. Households have to purchase prescription drugs, although these are often not available. Services become more expensive when referrals are made, for example to Kalango Missionary Hospital within Pader district. The main costs include admission fees, maternity costs, maintenance while in hospital, drugs and transport costs (due to the distance from Obolokome and Lira Palwo TRS). The district also lacks the human resource capacity to implement and manage its planned activities. The District Development Plan notes a staffing gap of 64% and even lacks the necessary revenue to meet the wage bills of its current staff. Therefore, while funds from the central government have been allocated towards different activities, the district can in some cases not afford the management costs associated with implementing projects and budget lines remain unspent. A survey of mortality and morbidity in Agago county (where Aywee Keyo is located) in 2007 has shown that the lack of drugs at health centres in return areas, the lack of water and hygiene services and household decisions not to seek health care because of the distance to betterequipped centres have contributed to higher mortality rates in return areas, including transit sites, particularly for children under the age of five (Medair Uganda, 2007). Local government capacity is particularly constrained by these factors at the sub-county levels. The system of leadership established in the main IDP camps continues to function alongside official local government structures, the Local Councils (LC), which the government has sought to re-establish as part of the resettlement process. Whereas law and order is supposed to be the purview of local councillors, some camp leaders have tended to arbitrate disputes within camps. In 2009, the government communicated to agencies that they were to engage with LC structures only, although all the structures are far from being in place; the district water department, for example, has only three staff for the whole district. Most children of primary school age are in school, and a majority of these children have a school uniform. However, a number of the poorest households in the study reported that one or more children did not own a uniform. Although official policy states that children without uniforms should not be barred from school, attendance of these children is likely to be affected both by the stigma of not wearing a uniform and by parents’ failure to pay ‘unofficial’ Parent Teacher Association (PTA) charges. Eight households in the study households have children in secondary school (not all households have children of secondary school age). In two of 5.1.2 Access to services for the study population 12 Until mid-2007 the study population benefited from an NGO soap distribution that was extremely popular as it meant that more money was available for other essential items or to access health, education and other services. Access to basic services has been very restricted for the majority and has thus had an impact on 23 these households, older brothers in the police service help with school fees; one household is helped by a son in the army, three households are assisted by uncles living away from Lira Palwo and one boy works as a labourer to pay his fees; in another household, an uncle pays for one child and the father pays for the second. The majority of these children are boys (two are girls). Over the past three years, one girl in secondary school became pregnant and dropped out; another died of measles. been slow and is being held back by the lack of access to veterinary services. Two households in the study group, for example, have lost oxen in 2009; one had benefited from NUSAF last year, and the other had bought the animal with dowry money. The loss of livestock is a huge setback for the households concerned. Restocking programmes have also been affected in 2009 by a countrywide ban on animal movement following a foot-and-mouth disease outbreak in the neighbouring district in 2008. The ban was still in force in April 2009. Within this study’s sample, there has been a slow but steady increase in the number of pigs and goats. Poultry numbers are more difficult to track as they can vary widely during the course of the year. For example, around half of all households in the sample had chickens in 2007–08; however, the rate fell dramatically in 2009 due to endemic disease. At the time of the last assessment, no households reported owning any chickens. 5.2 Humanitarian and development assistance Beyond food aid, the study population has received remarkably little assistance during the critical recovery period. The scale and coverage of livelihood interventions implemented by government and aid agencies have been limited and there is a clear gap in effective and appropriate assistance to support people in the recovery of livelihoods. NUSAF has helped some households to acquire cattle. However, the programme has been fraught with problems and many members of the community feel they have been unfairly excluded. Existing restocking programmes have benefited a limited number of households only, and problems with the allocation of NUSAF grants have been noted. 5.2.1 Agricultural assistance In 2008–09, some of the study population were among the beneficiaries of an agricultural rehabilitation project implemented by Mercy Corps. The objective was to increase production, with a focus on assisting the poorest. Assistance included tillage support (a voucher to pay for the hire of an ox plough to clear one acre of land) and a cash voucher for seed purchase. The tillage project was specifically designed for the labourpoor. The tillage scheme successfully targeted the poorest but accounted for just six acres (2.4 hectares) out of the total 101 acres (40.9 hectares) cultivated by the 23 households in the study during 2008–09. 5.3 Credit and Savings There are a number of credit and savings projects in northern Uganda, many of which started in the IDP camps. One such project, supporting the establishment of group savings funds and implemented by Care International, was used by about a quarter of the households in the study. Households indicated that, if loans were available, their first priority would be improving their draft power through the purchase of oxen, ox ploughs and other agricultural implements. Overall returns on the agricultural inputs programme were disappointing; in one case, the total value of input support (UGX 103,000 for combined tillage and seeds) was less than the value of the crop produced. This is not because the programme was poorly conceived or executed, but simply because the rains failed, highlighting the need to approach agricultural rehabilitation as a multi-year process. Donors and NGOs have identified lack of agricultural inputs as a problem, but the resources that have been made available for 2009 are woefully limited. In the study sample, only one household had direct access to loans. In 2009, this household accessed a loan from the Savings and Credit Co-operative (SACCO), which aims to support agricultural extension services. They have borrowed UGX 300,000 and will have to repay the loan in six months with a cumulative interest rate of 18%. SACCOs form a central part of the government’s Rural Finance Strategy, but they have been criticised for causing market distortions, with the Ugandan government seen as being too close to direct microfinance provision. As with other schemes, there have 5.2.2 Livestock support The process of restocking, either by individual purchase or through assistance programmes, has 24 been cases of fraud and corruption (Banfield and Naujoks, 2009). low, and a successful harvest is not guaranteed (LEMU, 2007). The majority of households produce so little that they are not involved in any market transactions and levels of disposable income have decreased further, particularly since the phase-out of food aid in December 2008. This is likely to further reduce involvement in the market as households prioritise their own consumption needs. Few people are interested in incurring debt to invest in agriculture: interest payments are too high, repayments must often be made within a very short timeframe—sometimes just one month, yet returns on agriculture take many months. Prices paid to farmers for agricultural produce are 25 6. Conclusions The findings of this study show that many factors contribute to the return of people from IDP settlements to their home and to recovery. Decisions around the timing of return for households in the study group, as for the wider population, were informed by the opening of schools, the availability of thatching grass for roofing, the availability of labour mainly during the dry season (December to March) to build houses, and a sense of security. Consistent concerns highlighted by households were: (i) the lack of health services and safe water in their home villages; (ii) robbery and theft, described by some respondents as ‘rampant’; (iii) a concern that they would not be able to afford to keep children in school when they reached the higher classes; and (iv) an underlying sense of insecurity due to ongoing fighting between the government and the LRA. livelihoods or invest in livestock (as savings as well as draft power). Food aid has been critical to resettlement; without it, people could not have survived for the first two years while reinstating their homes and agriculture. Now that food aid has ceased and other forms of assistance are extremely limited, livelihoods are likely to deteriorate. A recent FAO assessment found that the ending of food aid combined with poor rains and high food prices has led to a food crisis; FAO thus recommends resuming targeted food aid (FAO, 2009). Many agricultural rehabilitation projects focused on quick impact and increasing overall production over a wide area, but this has meant that assistance mainly benefited households that have already been able to clear their land, while those who lack labour were left out. The timeframe for reestablishing agricultural livelihoods is unpredictable and returns on investment rely as much on factors outside donor control (the weather, the price of oil, insecurity) as they do on good project management and design. This indicates the need for substantial investment to recapitalise production over a longer period than the normal 12–18-month emergency funding cycle. The weather and the security context are key factors in the rate of recovery and, under these circumstances, the premise that ‘self-sufficiency’ can be restored to agricultural communities within an 18-month aid cycle is simply inaccurate. The current situation of widespread food insecurity will create a further deterioration in the livelihoods and wellbeing of returnees. Access to schools has been a key factor in households’ decisions around when and where to move, to a far greater extent than access to health services. This underlines the value parents place on education, and the need for continuing donor commitment to the sector. In the TRS in Obolokome, some parents had started had started to collect money from their own communities to fund teacher training, due to the lack of government resources. In households with older children, the cost of fees for secondary school was also a major worry. Limited services not only posed longer-term risks to livelihoods (through lack of education), but also immediate risks to life, as mortality has been found to be higher in return sites than in the camps (Medair Uganda, 2007). Lack of services is clearly a key constraint to recovery. Increasing political instability in the north following the end of the peace talks in December, and government attacks on the LRA in DRC in the same month, means that there is a real risk of a deteriorating security situation and renewed conflict in the north. In addition, at the local level, the return process is accompanied by increasing conflict over land, and banditry and theft is rife. These factors pose serious risks to recovery, and indicate a need to retain capacity for humanitarian response for the foreseeable future. Determinants of recovery include the availability of household labour, skills and assets, and, in less well resourced households, access to assistance from other family members. While there was no shortage of land, a lack of inputs and labour was a major constraint on food production. In addition, income-earning opportunities were mainly limited to low-paying activities, such as agricultural labour, sale of grass and firewood, construction work and brick making. During times when households were involved in land clearance and house construction, their income dropped. As a consequence of these constraints, people have been unable to rapidly rebuild their former The findings from this study regarding household experiences of return, insecurity and highly vulnerable recovery processes raise questions around whether the conditions are in place for recovery-oriented programming to be implemented 27 without the continued provision of humanitarian assistance, and whether the withdrawal of food aid and the drastic reduction in humanitarian relief have contributed to making the outcome of recovery strategies increasingly uncertain. maintain the flexibility to respond to a possible resumption of conflict. A similar lack of adequate support to processes of livelihoods recovery was found in Southern Sudan (Pantuliano et al., 2008). By contrast, in Burundi refugees returning from Tanzania received relatively generous cash grants in addition to food aid, which played a valuable role in stimulating recovery and supporting investment (Haver et al., 2009). The paper suggests a need for caution about prematurely and abruptly cutting off humanitarian aid, a need for long-term social assistance for the most vulnerable and a need for more sustained and generous recovery and livelihood support based on local economic possibilities and constraints. There is also a need for better coverage of assistance and better access to basic services. This does not only mean better access to education, health care, and agricultural support but also requires a focus on improving security through greater attention to rule of law; for example by investing in police reform and ensuring they receive adequate salaries and training; as well as supporting the strengthening of judicial structures. There is also a need to Sadly, support for continued relief or for longerterm social protection and effective assistance to enable the recovery of livelihoods has been in short supply, limiting the extent to which poorer households have been able to recover. The ending of food aid and inadequacy of other assistance has left many people struggling to meet absolutely basic needs. This detailed picture of how a small number of households have fared since 2006 suggests a missed chance to seize the opportunity provided by relative peace and security to really kick-start recovery. 28 Appendix Individual household model definitions and terms (a) To compare the income and standard of living of different households, food and non-food income have been converted to an equivalent common measure: kilocalories. (b) To establish comparability between households of different size and composition, results have been standardised in terms of ‘adult equivalents’. (c) To compare household income, results are expressed in terms of ‘disposable income per adult equivalent’. Disposable income is the amount of income that would remain after the household has met only its basic food energy requirement. (d) To consider the extent to which households can meet their minimum set of needs, including both food and non-food essentials, a basic ‘standard of living’ threshold has been set. Households below the standard of living threshold do not have sufficient disposable income per adult equivalent to meet these basic needs. Methodology (i) Households in the sample obtain their income partly as food produced for consumption, partly as cash (from the sale of crops, remittances, gifts from relatives), partly as food aid and partly as wild food. Converting all income into its money equivalent runs into the difficulty that some items of food produced (e.g. fruits) do not have a ‘market price’: they cannot be sold in the volumes produced. To compare the relative contribution of these different sources of income, all income is converted into its approximate kilocalorie (kcal) equivalent. Thus, cash income is converted into the kcal maize (per adult equivalent) that could be purchased with the cash earned. (ii) To compare households of different sizes and with different demographic characteristics, results are expressed per adult equivalent. The number of adult equivalents is produced by weighting the requirement of each member of the household according to age and sex, based on WHO (1985) estimates of their different food energy requirements. For example, an adult male would be equal to one adult equivalent, while an eight-year-old girl would be equal to 0.83 adult equivalent. (iii) Disposable income: As food produced by the household and received as food aid is less than the amount required to meet basic needs, any unmet household food requirement is ‘purchased’ in the IHM model, at the price prevailing in the study period, and the cost of purchase is subtracted from the household’s money income. The amount remaining, when basic food needs have been met, is termed ‘disposable income’. (iv) A minimum standard of living has been defined as the ability of a household to meet: • basic household expenses, i.e. kerosene/candles; household utensils; salt; borehole fee; • personal expenses, i.e. clothing, soap and medical costs; and • primary school costs, including books, uniforms and school materials. These costs are allocated to each household, based on the number and age of household members. The cost of these items is based on local market prices. Calculations were carried out using Evidence for Development’s IHM software. 29 30 References Acidri, J. (2007) Livelihoods in Crisis Situations: 2009, UNHCR Policy Development and Evaluation Household Economy Assessment (HEA), Pader District Technical Report (unpublished). Service. HRW (Human Rights Watch) (2005) Uprooted and Forgotten: Impunity and Human Rights Abuses in Northern Uganda. Vol. 17, no. 12(A). Allen, T. (2005) War and Justice in Northern Uganda: An Assessment of the International Criminal Court’s Intervention—An Independent Report. Final draft. London: Crisis States Research Centre, Development Studies Institute, London School of Economics. International Alert (2008) ‘Building a Peace Economy in Northern Uganda’, in Investing in Peace, no. 1, September. London: International Alert. ASB (2008) ‘Monthly Sub-project Monitoring Report 07/SB/UGA/LS/438’, in Return Monitoring, Camp LEMU (Land and Equity Movement in Uganda) (2007) Linking Land Tenure and Agricultural Modernisation: Making PMA Relevant for All. Coordination, Camp Management, Camp Phase-out, and Participatory and Integrated Community Projects— Southern Pader District (Puranga, Awere, Omot, Patongo, Lira Palwo and Kilak). Lucima, O., ed. (2002) Protracted Conflict, Elusive Peace: Initiatives to End the Violence in Northern Uganda. Accord, iss. 11. Banfield, J. and J. Naujoks. (2009) Enabling Peace Economies through Early Recovery: Perspectives from Uganda. Strengthening the Economic Dimensions of Medair Uganda (2007) Mortality and Morbidity Survey: Agago County, Pader District, Northern Uganda. Peacebuilding Case Study Series. London: International Alert. MSF Holland (2004) Internally Displaced People in Lira and Pader, Northern Uganda: A Baseline Health Survey. Bailey, S., Pavanello, S., O’Callaghan S. and Elhawary. S (forthcoming). Untangling Early Recovery. HPG Discussion Paper. London: Overseas Development Institute. Preliminary report. Bradbury, M. (1999) An Overview of Initiatives for Peace in Acholi, Northern Uganda. Reflecting on Peace Pantuliano, S. et al. (2008) The Long Road Home: Opportunities and Obstacles to the Reintegration of IDPs and Refugees Returning to Southern Sudan and the Three Areas, HPG Synthesis Paper. London: Practice Project. Overseas Development Institute. Branch, A. (2008) ‘Against Humanitarian Impunity: Rethinking Responsibility for Displacement and Disaster in Northern Uganda’, in Journal of Intervention and Statebuilding, vol. 2, iss. 2, pp. 151–73. Petty, C. and K. Savage (2007) Livelihoods in Crisis: A Longitudinal Study in Pader, Uganda—Inception Report. Dolan, C. (2000) What Do You Remember? A Rough Guide to the War in Northern Uganda, 1986–2000. RLP (Refugee Law Project) (2007) What about Us? The Exclusion of Urban IDPs from Uganda’s IDP-related Policies and Interventions. Briefing Paper. Kampala: HPG Working Paper. London: Overseas Development Institute. COPE working paper 33. London: ACORD. RLP Dolan, C. et al. (2006) Humanitarian Protection in Uganda: A Trojan Horse? HPG Background Paper. London: Overseas Development Institute RLP (Refugee Law Project) (2008) Is the PRDP Politics as Usual? Update on the Implementation of Uganda’s Peace Recovery and Development Plan, Briefing Note FAO (Food and Agriculture Organization) (2006) Land no. 2. Kampala: RLP access and land use mapping in Northern Uganda Intermediate Report on Land Access Mapping RoU (Republic of Uganda) (2005) Health and Mortality Survey among Internally Displaced People in Gulu, Kitgum and Pader Districts, Northern Uganda. Kampala: FAO (2009) Rapid Food Security Assessment Report: Acholi and Teso Sub-regions, www.ugandaclusters.ug. Ministry of Health Finnstrom, S. (2008) Living with Bad Surroundings: War, History and Everyday Moments in Northern Uganda. Durham: Duke University Press. RoU (Republic of Uganda) (2007) The Peace Recovery Haver, K. et al. (2009) Evaluation of the Use of Cash RoU (Republic of Uganda) (2009) National Budget and Development Plan for Northern Uganda, 2007– 2010. Kampala: Republic of Uganda Framework Paper FY 2009/2010–FY 2013/2014. Grants in Voluntary Repatriation to Burundi, 2007– 31 Kampala: Ministry of Finance, Planning and Economic Development. Assistance Programme for Northern Uganda 2009– 20011 (UNPRAP), draft, July. Ruaudel, H. and A. Timpson. (2005) Northern Uganda: UNOCHA (2007a) Uganda Consolidated Appeal 2008, accessed 14 April 2009, www.reliefweb.int From a Forgotten to an Unforgivable Crisis—The War against Children. Situation Report. Institute for Security UNOCHA (2007b) Waiting for Godot in Gulu: Possible Reasons for Delay in IDP’s Return Process. Studies. Savage, K., C. Petty and J. Acidri (2008) Livelihoods in UNOCHA (2009a) Uganda Consolidated Appeal 2009, Mid-Year Review. Crisis: A Longitudinal Study in Pader, Uganda—Year Two Update. HPG Working Paper. London: Overseas Development Institute. UNOCHA (2009b) Pader District IDP Camps and Return Sites Population. Uganda Clusters Map Centre, 8 July, Stites, L., D. Mazurana and K. Carlson (2006) Movement on the Margins: Livelihoods and Security in Kitgum District, Northern Uganda. Medford, MA: Feinstein www.ugandaclusters.ug. International Center, Tufts University. UNOCHA (n.d.) Financial Tracking System, accessed 28 April 2009, www.reliefweb.int/fts. UN (2008) Annual Report of the United Nations High Commissioner for Human Rights and Reports of the Office of the High Commissioner and the SecretaryGeneral. A/HRC/7/38/Add.2. New York: UN General WFP (World Food Programme) (2009) Country Strategy for WFP in Uganda (2009–2014). WHO (World Health Organisation) (1985) Energy and protein requirements for the population of a typical developing country. WHO technical report series 724, Assembly. UNCT Uganda (United Nations Country Team Uganda) (2009) United Nations Peace Building and Recovery Geneva 32 Overseas Development Institute 111 Westminster Bridge Road London SE1 7JD UK Tel: +44 (0)20 7922 0300 Fax: +44 (0)20 7922 0399 Email: [email protected] Website: www.odi.org.uk