Content uploaded by Catherine-Lune Grayson
Author content
All content in this area was uploaded by Catherine-Lune Grayson on Mar 17, 2016
Content may be subject to copyright.
DURABLE SOLUTIONS | 1
DURABLE SOLUTIONS
Perspectives of Somali Refugees Living in Kenyan and
Ethiopian Camps and Selected Communities of Return
DURABLE SOLUTIONS | 3
Acknowledgements
The Durable Solutions report is a research project designed by Catherine-Lune Grayson
and carried out in collaboration with research interns Andre Epstein and Emily Coles,
and three Somali organisations: the Sean Deveruxe Human Rights Organisation, the
Somali Children Welfare and Rights Watch in Baidoa and the Gedo Social
Development Organisation. David Guillemois oversaw the compilation and analysis
of the data.
Durable Solutions research has been conducted with support from the Danish Refugee
Council’s Regional Office in Nairobi and its field offices in Dadaab, Kenya, Jijiga and
Dolo Ado, Ethiopia, as well as the Norwegian Refugee Council in Kakuma, Kenya.
Special thanks to UN agencies, non-governmental organisations, civil society
organisations and local authorities who shared their views on durable solutions
for Somali refugees, and to all refugees and returnees who shared their time and
perspectives with us.
Funding for this initiative was generously provided by the Swiss Agency for
Development and Cooperation.
Design and layout: Alexandra Strand Holm | DRC & Jean Bedell Omboke | Willart
Cover photos: Alexandra Strand Holm
4 | DURABLE SOLUTIONS
ARRA Administration for Refugee and Returnee Affairs (GoE)
CBO Cross-Border Operation
DRA Department of Refugee Affairs (GoK)
DRC Danish Refugee Council
FGD Focus Group Discussion
FSNAU Food Security and Nutrition Analysis Unit – Somalia
GoE Government of Ethiopia
GoK Government of Kenya
GSDO Gedo Social Development Organisation
HH Household
HRW Human Rights Watch
ICU Islamist Courts Union
IDP Internally Displaced Person
ITW Interview
NGO Non-governmental organisation
NRC Norwegian Refugee Council
OAU Organisation of African Unity
OCHA Office for the Coordination of Humanitarian Affairs (UN)
PRM American Bureau of Population, Refugees and Migration
QIPs Quick Impact Projects
SCWRW Somali Children Welfare and Rights Watch
Sedhuro Sean Deveruxe Human Rights Organisation
SNM Somali National Movement
TFG Transitional Federal Government
UN United Nations
UNHCR United Nations High Commissioner for Refugees
US United States
USD United States Dollars
WASH Water, Sanitation and Hygiene
Acronyms
DURABLE SOLUTIONS | 5
Executive Summary........................................................................................................
1 Preliminary remarks and purpose of the research.........................................................
2 Methodology and structure of the report......................................................................
3 Somali Displacement...................................................................................................
4 Repatriation efforts to Somaliland and Puntland from Ethiopia......................................
a) Repatriation from eastern Ethiopia....................................................................
b) Observations .............................................................................................
5 Somali refugees in Kenya and Ethiopia.........................................................................
a) Overview of legal frameworks and government policies.....................................
b) Enhanced self-reliance and local integration......................................................
c) Resettlement....................................................................................................
6 Refugees in Kenyan and Ethiopian camps....................................................................
a) Overview of the camps..........................................................................................
1. Kenya..............................................................................................................
2. Ethiopia...........................................................................................................
3. Place of origin and ethnicity..............................................................................
4. Ethnicity...........................................................................................................
5. Gender balance...............................................................................................
b) Presentation and analysis of the findings..........................................................
1. Profile of households interviewed.....................................................................
2. Livelihood activities..........................................................................................
3. Education and Vocational training.....................................................................
4. Safety..............................................................................................................
5. Intentions and pre-conditions for repatriation....................................................
6. Decision-making and sources of information....................................................
7. Observations...................................................................................................
7 Perspectives of selected communities of return............................................................
a) Overview of the situation in selected districts of return............................................
1. Afmadow in Lower Juba Region.................................................................
2. Baidoa in Bay Region.........................................................................................
3. Belet Xaawo in Gedo Region...............................................................................
b) Impact of displacement and of an eventual large-scale return of
refugees and IDPs............................................................................................
1. Spontaneous returns of refugees and IDPs............................................................
2. Impact and pre-conditions for a sustainable large-scale return..........................
3. Observations .............................................................................................
8 Concluding remarks.....................................................................................................
Bibliography....................................................................................................................
Annex A: Methodology....................................................................................................
Annex B: Interview planning............................................................................................
Table of contents 7
12
15
17
21
21
22
25
25
25
26
28
28
28
29
29
30
30
30
30
31
33
35
35
38
39
41
41
41
42
44
45
45
45
46
49
50
54
55
6 | DURABLE SOLUTIONS
DURABLE SOLUTIONS | 7
Executive Summary
Continued crises of governance in Somalia since the early 1990’s have provoked several episodes of large-scale internal and external
displacement (Menkhaus, 2008: 4, 21). More than a quarter of Somalia’s population has been forced to ee. As commonly observed in
large-scale displacement situations, the majority has moved to a safer area within the country, but large numbers have also ed to another
country. As of August 2012, some 1.36 million people were displaced within Somalia, and more than a million, mostly from south-central
Somalia, had claimed asylum in another country (UNHCR Information portal, 13 Sept. 2012; UNDESA, 2011: 31).
A signicant number of Somali refugees have been living in Kenyan and Ethiopian camps for up to two decades. Encampment policies
limiting refugees’ freedom of movement and access to labour markets have undermined the quality of the asylum offered in both countries.
Camps, usually considered temporary, have turned into a de facto long-term solution. In Kenya, nearly half a million refugees live in Dadaab
and Kakuma camps. In Ethiopia, camps in the Somali region, near Jijiga and Dolo Ado, host some 211,000 refugees (UNHCR Information
portal, 13 Sept. 2012).
Repatriation, local integration and resettlement prospects are limited for refugees in camps1. Large-scale formal local integration is improb-
able for politico-historical reasons and resettlement is only offered to a small proportion of refugees. Repatriation, which is not necessarily
the optimal solution for all, is likely to be the only option available to the vast majority. Yet, conditions in Somalia are not currently conducive
for a sustainable return. While there has been political change in the situation in south-central Somalia in recent months, notably with
presidential elections in August 2012, stability and safety remain uncertain and their lasting character will only be established with time.
Previous episodes of repatriation to Somalia and Somaliland from eastern Ethiopia and northern Kenya have shown that lasting security,
basic services, and humanitarian access are minimal prerequisites for successful and durable repatriation and reintegration. Careful attention
to livelihoods and access to land and their intersection with clan territory and inter-clan relations are essential, as is the participation of clan
leaders in planning and mobilization. Experience suggests that smaller and incremental population movements, coupled with assistance
and monitoring of returnees can help make return and reintegration sustainable. It also highlights the importance of viewing repatriation
as more than simply a logistical movement of people, but rather an integrated part of wider development and peacebuilding projects.
This research was undertaken as an attempt to better understand how displaced Somalis view their situation, envisage their future, and
if their repatriation ever became possible, what would make it sustainable. To do so, several hundred refugees were interviewed and
participated in Focus Group Discussions (FGD) in Kenyan and Ethiopian camps. The perspective of three selected communities in areas
of origin or anticipated return was also explored through interviews with key informants in Afmadow in Lower Juba, Baidoa in Bay and
Belet Xaawo in Gedo.
Given its relatively small sample, the research was not designed to allow statistical projections, but to permit pertinent observations. It is
not a planning document, but an analytical input aimed at contributing to further planning on durable solutions.
The following pages present a summary of key ndings.
1 UNHCR acknowledges voluntary repatriation, local integration and resettlement as the three possible durable solutions, i.e., solutions allowing refugees to secure the political, economic, legal
and social conditions needed to maintain life, livelihood and dignity.
8 | DURABLE SOLUTIONS
Return intentions and reintegration prospects
Given the choice, nearly half of Somali refugees interviewed in camps would prefer to move to a third country, a third would return to
Somalia and small numbers would integrate locally. Yet, if conditions were conducive in Somalia, more than half would consider returning.
In light of their previous experience, refugees recurrently stress that improvements in stability and security would have to be long-lasting
before they opt to return. A number of refugees in camps have, indeed, experienced several episodes of displacement, due to short-lived
improvements in security and stability, and would do their utmost to avoid being displaced yet again. However, local integration did not
emerge as a popular option, possibly because refugees judge their conditions in asylum difcult as well as recognising that local integra-
tion is not an option recognised by the Government of Kenya.
As the duration of exile increases, repatriation and reintegration prospects decrease, even if conditions for return became conducive.
Approximately half the refugees interviewed in the study had access to land in Somalia before eeing. Of that number, slightly less than half
said they would still have access to it, a fth did not know. A signicant difference exists between refugees who arrived before 2006 (40%
say that they had access to land and less than half of them believe they would still have access to it) and those who arrived in 2011-12 (80%
had access to land, with less than half believing they would still have access to it and an important proportion saying they did not know).
The fact that many youths have spent most of their lives in exile and have limited recollection and personal connections with Somalia,
if any, might come into play. For youths who grew up in camps, which can, to some extent, be compared to cities (cf. Jansen, 2011),
returning to the rural areas many come from, with limited public facilities, seems especially difcult to envisage. For different reasons, the
return and reintegration of people who have had traumatic experiences or experienced direct violence also appears unlikely, regardless of
the duration of their exile or their bond with Somalia. It is clear that no response will be adequate for all refugees, thus there is a need to
be creative in the search for durable solutions and further explore pathways to local integration and mobility opportunities.
Nearly all refugees mentioned ghting as one of the drivers of their displacement. Insecurity was also commonly raised, while small
numbers cited drought, loss of livestock and loss of livelihood. A slightly greater proportion of refugees intending to return listed drought,
livestock depletion and loss of livelihood as some of the reasons for their displacement. Although insecurity may be the overriding factor
for displacement, those arriving in 2011 were affected by the drought .
The length of exile, drivers of displacement, age, belonging to a minority group and conditions in the country of asylum appear as impor-
tant factors when examining return intentions and reintegration prospects of Somali refugees. Having been displaced before, receiving
remittances or having beneted from education or vocational training classes did not seem to signicantly inuence return intentions.
Pre-conditions for returning
Before taking the decision to return, in addition to considering security and safety, refugees would assess access to education and health
care, their livelihood opportunities and whether humanitarian assistance would be available upon return. Only a small number of refugees
said that if they were to repatriate, they would return to a different location than their locality of origin, usually an urban area for livelihood
opportunities, the urban way of life, security and access to basic services. In all locations, refugees stressed that conditions were harsh
and could induce premature returns.
In all locations except Jijiga, only a minority reported no longer having relatives in Somalia. In general, the proportion of people with
no known relatives in Somalia increased with the duration of exile. Very few people knew refugees who had repatriated since they had
arrived in the camps. Those who did mostly explained such returns by the difcult living conditions and lack of livelihood opportuni-
ties in the camps, the improvement of security in their country, family bonds in Somalia, and the need to resume agricultural activities.
Enhancing self-reliance in camps
At this point in time, a majority of refugees in camps assess that the reasons that led to their displacement still prevail, and thus would
not envision returning until conditions improve. In the meantime, in addition to ensuring that standards of living and security in camps
are and remain satisfactory, refugees note that aid organisations could take measures in the camps to enhance their reintegration
prospects through skills building, i.e., by ensuring greater access to education, including for adults, vocational training - notably in
tailoring, business, carpentry, driving, computer literacy, language, electricity and mechanics - and to livelihood opportunities. Refugees
hired as incentive workers stressed they could benet from advanced on-the-job training, skills transfer and mentoring, that could lead
to enhanced responsibilities. Key informants in areas of origin also regularly emphasised that the return of refugees with education and
training would be helpful locally.
The perspective of refugees in camps
DURABLE SOLUTIONS | 9
For the time being, educational and livelihood opportunities in the camps remain relatively limited, even though efforts have been made
to boost access to such opportunities. One third of the refugees interviewed said they had gained useful skills in exile, most commonly
education, business skills and tailoring. Refugees who have been in camps longer and men generally seem to have more prospects
in camps as they have relatively better access to livelihood opportunities and are more mobile compared to women who have cultural
constraints and responsibilities. Similarly, more established camps offer better opportunities than newly opened ones.
Approximately 45% of households interviewed pursue at least one livelihood activity in the camps, most commonly petty trade, small
shops or casual labour. Adults under the age of 35 appear more likely to have a livelihood activity than older people. Education in the
camps tends to be insufcient. Existing facilities are in poor condition and under-equipped, while teachers are under-qualied. Enrol-
ment, attendance and levels of retention are low, especially for girls (Dippo, 2011: 19; UNHCR et al., 2012: 9, 15 and 25; UNHCR,
2011a: 31). Over two thirds of refugees that arrived before 2006 said at least one of their family members had been to school. This
proportion dropped to less than half for those who arrived in 2011-12. Finally, the proportion of refugees who have attended vocational
training classes is low. Refugee leaders, entrepreneurs and employees have had slightly more access to training, while opportunities
for women and vulnerable refugees seem limited.
Assisting repatriation and reintegration
Refugees highlighted that when conditions become conducive for return - in their words: when there is signicant progress in peace,
justice, security and stability -, humanitarian organisations should be ready to support repatriation and reintegration, through transport
and basic assistance, and to assist refugees with access to land and housing. If refugees were repatriating, most plan to farm, raise
livestock or pursue business activities, which makes access to land in areas of return crucial. Prior to eeing, nearly half of the refugees
interviewed farmed and/or raised livestock.
The perspectives of selected communities of return
The situation in Somalia differs from one location to another. While Somaliland and Puntland are relatively stable and conict-free, most
of the south-central region remains volatile, with limited humanitarian access and continued population displacement.
A majority of Somali refugees in Kenyan and Ethiopian camps ed from the south and central regions. Even though the security
landscape in the area has been changing rapidly since the beginning of the year, there is no evidence yet of a signicant improvement
in the generally poor human rights situation for the local population (HRW, July 2012).
The three localities studied, Afmadow in Lower Juba, Baidoa in Bay and Belet Xaawo in Gedo, were recently captured from Al-Shabab.
The three regions host signicant populations of internally displaced persons (IDPs) (31,000 in Lower Juba, 40,000 in Baidoa and
77,000 in Gedo) and are the areas of origin of a signicant number of refugees (more than 200,000 are from Lower Juba, incl. more
than 48,000 from Afmadow, 80,000 from Baidoa, 124,000 from Gedo).
In the three locations, similar to most of south-central Somalia, after two decades of conict and instability and in the absence of a
functioning government, public infrastructure and basic social services are very limited and, when available, are frequently of unsatisfac-
tory quality. In the districts studied, small numbers of refugees and IDPs are reported to have returned spontaneously.
Pre-conditions for sustainable and peaceful returns
Authorities in Kenya have discussed the creation of “Jubbaland” in southern Somalia as a safe zone intended to curb the ow of refugees
into Kenya and allow for the repatriation of refugees to southern Somalia. Without clear benchmarks to ensure that sustainable security
is in place in areas of return, that essential services are available, and that humanitarian organisations have secure access, a repatriation
programme would likely be unsustainable and create tensions and competition over scarce resources.
So as to make large-scale returns possible and sustainable, key informants in all districts pointed out that careful planning was essential
and that hasty repatriation would risk stability and peacebuilding. Planning should include access to land and livelihoods, quality basic
services, shelter, food and water.
10 | DURABLE SOLUTIONS
Struggle for land has been a central element to the Somali conict. In order to avoid creating further conicts, a fair and efcient land resti-
tution system, addressing restoration and compensation, will be necessary. Indeed, access to land will be a crucial and sensitive matter in
both rural and urban areas. Although a large part of the potentially repatriating refugee population will rely on farming and agro-pastoralism
for their livelihood, hence requiring signicant land access, those returning to urban areas will require assistance in terms of housing, land
and property matters. The situation for women and minorities will require special attention, as certain customary rules discriminate against
them, which could further limit their reintegration prospects. It is recognised that particularities exist for urban areas which would require
further research beyond the scope of this study.
Key informants also suggested that refugees should be equipped with practical skills while in exile. In all districts, a critical shortage of
qualied staff was highlighted, notably due to large-scale departures - a number ed, while some of those who remained, lost their jobs,
as schools and health centres were left unused, and ended up seeking job opportunities elsewhere. Unattractive wages and a lack of
training structures have also encouraged a number of qualied people to move to Kenya or Ethiopia.
The dearth of teachers, doctors and other health professionals is deemed especially critical. The need for administrators was also commonly
highlighted, while the need for engineers and business people capable of contributing to the development of the local economy also
emerged. In general, people with skills were described as in high demand. Returnees with assets, such as farming tools, livestock and
sewing machines, would have an easier time earning their living again, which, in addition to contributing to the local economy, would
facilitate their reintegration.
Several key informants stressed that humanitarian organisations could help communities prepare for sustainable repatriation by contrib-
uting to improving quality and access to basic services, building infrastructure (notably roads and electricity), enhancing food security,
supporting livelihoods and skills training and the local economy, as well as activities aimed at restoring and consolidating peace and
security. In addition to providing direct support to people returning in the form of food and shelters, aid organisations could support the
creation of mechanisms aimed at preventing and addressing land-related tensions.
Given the socio-economic situation of the population in Somalia, activities and aid targeting returnees should be balanced against support
to the overall population to ensure effective integration opportunities and avoid creating tensions and discrimination against returnees.
DURABLE SOLUTIONS | 11
12 | DURABLE SOLUTIONS
“If it is true that camps save lives in the emergency phase, it is also true that, as the years go by, they progressively waste these same
lives. A refugee may be able to receive assistance, but is prevented from enjoying those rights – for example, to freedom of movement,
employment, and in some cases, education – that would enable him or her to become a productive member of a society.”
UNHCR, 2004: 3
A signicant number of Somali refugees have been living in Kenya and Ethiopia for up to two decades, with limited prospects to repatriate,
integrate locally or be resettled. Encampment policies that restrict refugees’ freedom of movement and access to labour markets have
undermined the quality of the asylum offered
2
. Camps, usually considered a temporary solution, have turned into a de facto long-term
solution for Somali refugees. Since the organised repatriation from Ethiopia and Kenya from 1993 to the early 2000’s, only minimal research
and longer-term planning for durable solutions for Somali refugees has been undertaken.
UNHCR acknowledges voluntary repatriation, local integration and resettlement as the three possible durable solutions, i.e., solutions
that enable refugees to secure the political, economic, legal and social conditions needed to maintain life, livelihood and dignity. Recent
discussions have highlighted the need to be more creative in the search for durable solutions, notably through considering intermediate
integration3 and mobility within and between borders, as migration might help a number of refugees regain physical, legal and livelihood
security (Long, 2010; Lindley, 2011; UNHCR, 2012).
Repatriation has become the preferred durable solution since the 1980’s, notably due to the unwillingness of host countries to offer local
integration to refugees due to their inability to cater for these people in addition to their own national poor – and a sedentarist bias, according
to a number of scholars, e.g. Malkki (1995: 508) and Verdirame (2005: 335) –, is likely to be the only option offered to a vast majority of
Somali refugees. For politico-historical reasons, formal local integration is improbable for large numbers of refugees in Kenya and Ethiopia.
However, gradual approaches to local integration should not be excluded, notably for groups unlikely to repatriate voluntarily. While it can
be argued that refugee-hosting countries shoulder a disproportionate part of the burden and that greater resettlement efforts should be
made, resettlement remains unlikely to become an option for a signicant proportion of refugees.
“Local integration and repatriation are sensitive topics in the Kenyan context characterised by growing political support for the quick return
of refugees.” Authorities have discussed the creation of a safe zone inside Somalia, “Jubbaland”, to both stem the arrival of new refugees
and allow the repatriation of others (HRW, March 2012; Lindley, 2011: 29-30). Such a plan, that evokes the unsuccessful 1992 Cross-
Border Operation, would most likely lead to involuntary returns, and thus fail to respect recognised standards (Hyndman, 1998: 24-27)4.
Conditions in Somalia are not currently conducive for a large-scale sustainable repatriation: in late August 2012, more than a hundred
new arrivals from Somalia, mostly from Gedo region, were registered in Dolo Ado camps daily (ITW UNHCR Dolo Ado, 28 Aug. 2012). For
the majority of refugees, without a major and durable political solution, organised mass voluntary return is improbable in the foreseeable
future (Lindley, 2011: 4). Indeed, while there has been political change in south-central Somalia in recent months, notably with presidential
elections in August 2012, stability and safety remain uncertain and their lasting character will only be established with time. Forced returns
would be indefensible. Likewise, refugees should not be driven to repatriate by inadequate or deteriorating security and living conditions
in the refugee camps, but should return when they deem conditions conducive in their country of origin5.
Repatriation is not tantamount to mere physical return. It entails the restoration of meaningful citizenship, in which fundamental human
rights are protected by the state or, in the context of a fragile state, perhaps communities (Long, 2010: 3, 40). In such an environment,
repatriation needs to be considered and planned carefully, in conjunction with a broader peacebuilding and development strategy.
2 Long (2011: 8-10) notes that over the last two decades, the quality of asylum protection has been deteriorating, not only in Africa and Asia, but also in Europe. The focus on durable solutions
(and hasty and unsustainable repatriation) can notably be explained by the inadequate protection offered during displacement.
3 This report uses the definition and indicators of intermediate integration suggested by Banki (2004: 2-3). It refers to the “ability of the refugee to participate with relative freedom in the economic
and communal life of the host region.” It might entail partial self-sufficiency rather than full self-sufficiency. Unlike local integration, it might not include cultural and political participation, as well
as full legal rights. High levels of refugee integration are characterised by the fact that refugees are not restricted in their movements, own or have access to land, can participate in the local
economy, are moving towards self-sufficiency, are able to utilise local services such as health facilities and schools and are dispersed among the population.
4 In 1992, in response to pressures from the Government of Kenya, UNHCR attempted to create a preventive zone in southern Somalia to encourage refugees in Kenya to repatriate (Hyndman,
2000: 21).
5 Some 5,200 refugees returned to Somalia by their own means in March and April 2012, according to UNHCR’s Somalia population movement tracking (PMT). It is not known whether such
movements were temporary or permanent. Refugees returning to Somalia have cited the resumption of agricultural activities, but also reduced services and lack of livelihood opportunities in
the camps to explain their decision (UNHCR, April 2012: 2). Similarly, while some refugees in Dolo Ado in Ethiopia temporarily returned to harvest in the first months of 2012, according to refugees
interviewed in the camps in August 2012, a number of refugees also returned to Somalia permanently, due to the harsh living conditions in the camps (ITW UNHCR Dolo Ado, 28 Aug. 2012; HH
ITW and FGD, Aug. 2012). It has been repeatedly noted that a large number of refugees repatriated from Kenya to Somalia in 1993-94 because of the insecurity and deteriorating conditions in
the camps, and not because conditions were conducive in Somalia (Waldron, 1995: 13-15; Collin, 1996: 89-97).
1 Preliminary remarks and purpose of the research
DURABLE SOLUTIONS | 13
By putting institutions and communities under pressure, premature mass returns can hinder peacebuilding, reconstruction and sustain-
able reintegration. The departure of refugees is also likely to have a signicant impact on host communities, as camps have become an
intrinsic part of their economic survival and development6.
Moreover, repatriation is not necessarily the optimal solution for all refugees, especially in protracted situations (Kibreab, 1999: 390)
7
. Repat-
riating refugees may face serious obstacles (re)integrating into the socio-economic environment of their country of origin. Compounding this,
the return and reintegration process often takes place in difcult environments with derelict infrastructure and social services (Jacobsen,
2005: 9-10; Omata, 2012: 1). After decades in exile, it is unlikely that all refugees choose to return to Somalia, even if the reasons for
their ight (or their parents’ ight) cease to exist. Repatriation might entail signicant losses, notably in terms of livelihood, educational
opportunities and access to basic services (Jacobsen, 2005: 10; Long, 2010: 27).
Purpose of the research
This research was undertaken as an attempt to better understand how people who have been displaced due to conict and violence,
as well as recurrent drought and oods, view their situation, envisage their future and, if their repatriation ever becomes possible, what
would make it sustainable.
This project aimed to examine short and long-term opportunities for durable solutions, with a focus on repatriation, from the perspective
of refugees in camps8 and selected return communities9, with the objective of informing humanitarian organisations’ strategic planning.
While the immediate focus on repatriation might need to be carefully considered, it is likely to be the only option offered to a signicant
number of refugees. Thus, understanding how to better prepare refugees to reintegrate in a sustainable manner in Somalia is prioritised
for practical reasons. In addition, designing programmes to better prepare refugees for an eventual return, when conditions are conducive
for repatriation, should strengthen refugees’ ability to integrate locally or in a third country.
Repatriation and reintegration affect not only refugees, but also communities of return or of origin and refugee hosting communities. To
limit the scope and duration of the study, focus on the impact of the potential return of refugees on selected communities of origin was
chosen. Their absorption capacity and views on anticipated positive and negative impacts of such repatriation, notably in terms of access
to land and property, as well as skills and assets with which refugees could return to contribute to community recovery and reconstruc-
tion, were examined.
6 A 2010 assessment of the socio-economic and environmental impact of Dadaab refugee camps on host communities concluded that the total economic benefit of the camps and related
operations for the host community were around USD 14 million annually (Enghoff, 2010: 8-10).
7 The will to prevent local integration and ensure early repatriation means that refugees in Africa are treated as “temporary guests ” with curtailed rights, regardless of the length of their exile
(Kibreab, 1999: 390, 399). Such determination has also turned refugee camps, built around the notion of temporariness, into the preferred response to a refugee crisis, even though many
refugee situations last well beyond the emergency period and repatriation might not happen for years (Crisp, 1998: 28 and 2003: 3-4; Jacobsen, 2005: 108). This can also be linked to the shift
to preventive protection that focuses on the right to remain in one’s country rather than the right to seek asylum (Chimni, 1993: 443-444; Hyndman, 2000: 17).
8 While it would have been relevant and interesting to also include the perspective of refugees living outside camps, for capacity reasons, it was decided to focus on refugee in camps.
9 Return communities are understood as communities to which significant numbers of Somali refugees would potentially return if conditions became conducive for repatriation, based on
number of refugees originating from these communities.
14 | DURABLE SOLUTIONS
DURABLE SOLUTIONS | 15
In attempting to understand durable solutions from the perspective of refugees and return communities, some of the key elements that
had to be considered were: refugees’ views of their situation, and their potential for repatriation or local integration; refugees’ coping
mechanisms and livelihood practices in exile and their transferability; refugees’ relationship to their areas or country of origin and access to
assets left behind, including land, housing and other property; refugees’ (re)integration prospects in view of their education and livelihood,
knowledge of and relationship to their country of origin or of asylum.
The research focused on refugees living in camps in Kenya (Dadaab and Kakuma) and Ethiopia (Dolo Ado and Jijiga), as well as on selected
areas of origin or anticipated return: Afmadow in Lower Juba, Baidoa in Bay, and Belet Xaavo in Gedo.
Accurate statistical data on Somali refugees living in camps in Kenya and Ethiopia is scarce, notably with respect to their livelihood prior
to being in exile and while in the camps, assets in the camps and in their locality of origin, levels of education achieved in the camps, and
intentions in terms of durable solutions. This project did not aim to ll this gap, as a comprehensive census or registration exercise would
be necessary to gather this information. It was understood that the research would not allow statistical projections, but should permit
pertinent observations.
The methodology for the eldwork mixed qualitative and quantitative tools and approaches. Given the relatively small sample used in
refugee camps, it was decided to combine a number of household interviews with focus group discussions (FGDs), in order to corroborate
ndings and observations. Some 62 FGDs were held with a total of 410 refugees drawn from diverse populations in the camps, i.e., youth,
women, minority groups, people considered vulnerable (single-headed households, people with disabilities, those aficted by chronic
disease, the destitute, etc.), entrepreneurs, incentive workers, unemployed adults, recent arrivals (2011 and onwards), arrivals from 2006
and onwards, refugees who have been there for over a decade, camp leaders, including women representatives and elders. Interviews
were conducted with 360 randomly selected heads of refugee households. It was decided to interview heads of household rather than
individuals to elicit more information about other members of the family. Some 160 interviews took place in Dadaab, 40 in Kakuma, 80 in
Dolo Ado and 80 in Jijiga (cf. Annex A for details).
It was expected that by collecting data in all four main camp locations in Kenya and Ethiopia (Dadaab, Kakuma and Jijiga camps, and
four out of ve Dolo Ado camps), the sample would include refugees from all periods of arrival, from rural and urban areas, with differing
levels of wealth, age, gender and clan afliation, which would thus help capture the specicity of individual experience and determine
which factors are likely to inuence refugees’ intentions and views on the future. For example, while Ifo’s old residents mostly came from
rural areas, many of Hagadera’s old residents came from urban areas (PRM, 2011: 19).
Seventy-three interviews with key informants were conducted in three potential districts of return. The districts were selected based on the
areas of origin of refugees in camps, given that a number of refugees are likely repatriate to their area of origin, and based on humanitarian
access. In every district, interviews with key informants were conducted by Somali partners: the Sean Deveruxe Human Rights Organisation
in Afmadow (Lower Juba), the Somali Children Welfare and Rights Watch in Baidoa (Bay), and the Gedo Social Development Organisation
in Belet Xaawo (Gedo). Key informants included community leaders, elders, clan leaders, local authorities, service providers (health and
education), entrepreneurs, and civil society organisations.
Structure of the report
After briey outlining key episodes of Somali displacement and lessons learned from previous repatriation to Somaliland from Ethiopia, this
report presents the condition of Somali refugees in camps and their perspective on durable solutions. It then briey sketches the situation in
selected communities of return and introduces the perspectives of key informants on the potential large-scale return of refugees and IDPs.
2 Methodology and structure of the report
16 | DURABLE SOLUTIONS
DURABLE SOLUTIONS | 17
Continued crises of governance in Somalia since the early 1990’s have provoked several episodes of large-scale internal and external
displacement (Menkhaus, 2008: 4, 21), turning Somalia into “a country in exile” (Farah, N., 1998: 715). Approximately a quarter of all
Somalis, or over 2.3 million people out of a population of 9.4 million people, have been displaced (UNHCR Information portal, 13 Sept.
2012; UNDESA, 2011: 31). As commonly observed in large-scale displacement situations, the majority have moved to a safer area within
the country (thus categorised as internally displaced persons), but large numbers have also ed to another country (Jacobsen, 2005: 1
10
). As
of August 2012, some 1.36 million people were displaced within Somalia, and more than a million had claimed asylum in other countries,
primarily Kenya, Yemen and Ethiopia (UNHCR Information portal, 13 Sept. 2012)11. The Gulf States, Western Europe and North America
are also home to signicant Somali refugee populations (Jureidini, 2010: 5)12.
Movement has been driven by a complex combination of clan-based conict that has led to massive destruction of infrastructure and
livelihoods, and recurrent oods and drought. Although movement has been ongoing since the late 1990’s, three main periods of large-
scale displacement can be roughly distinguished:
i. Civil war, state collapse, clan war and drought (1988-1992): The Somali civil war that began in 1988 led to the collapse of the Somali
state and the fall of Siyad Barre’s regime in 1991. The violence, compounded by a severe drought in 1991-92, led to an acute humani-
tarian crisis that prompted large-scale displacement. Some 800,000 Somalis left for neighbouring countries. More than 600,000 people
from northwestern Somalia (now Somaliland) ed to the Somali Region of Ethiopia and Djibouti between 1988 and 1991. Clan-based
ghting in south-central Somalia between December 1991 and March 1992 displaced some 3.5 million people. Of that number, nearly
300,000 crossed the border to Kenya (Bradbury, 2010: 10; Gundel, 2002: 264; UNHCR Statistical Online Population Database).
While most of those who ed from Somaliland to Ethiopia and Djibouti repatriated between 1997 and 2005, only a relatively small
number of refugees in Kenya returned to Somalia13. In 2000, some 137,200 Somali refugees were still in Kenya, most of them in
Dadaab camps (UNHCR, 2002). No signicant movement of return has occurred since and only a minor proportion of refugees has
been resettled. UNHCR reports that some 10,000 children currently in Dadaab camps were born of parents themselves born in the
camps (UNHCR, 2012a).
ii. Ousting of the Islamic Courts Union (ICU) and ensuing violence (2006-2012): In late 2006, the ousting of the ICU, which had expanded
its control over most of southern Somalia, by the Transitional Federal Government (TFG), backed by the Ethiopian army, led to renewed
widespread violence. Fighting between clan militia and radicalised splinter groups of the ICU, notably Al-Shabab, and the Ethiopian
and TFG troops led to a massive humanitarian crisis and forced hundreds of thousands of people to ee (Menkhaus, 2010: 12-13).
More than 150,000 refugees arrived in Dadaab between 2007 and 2010. In Ethiopia, four new refugee camps were opened to accom-
modate new arrivals between 2007 and 2010. The number of Somali refugees in the country increased from 25,800 to 81,200 during
the same period (UNHCR statistics, June 2012; UNHCR Statistical Online Population Database).
iii. Drought, food crisis and lack of humanitarian assistance (2011-2012): In 2011, a severe drought hit southern Somalia. In the absence
of humanitarian assistance due to the insecurity preventing access, large numbers of agro-pastoralists and farming families who lost
their livelihoods were forced to move.
Some 164,000 new refugees arrived in Dadaab camps. In Ethiopia, three new camps were opened to receive some 98,000 new
refugees (UNHCR, 2011).
10 Jacobsen (2005) estimates that a third of those forcibly displaced by a conflict cross an international border. The majority will be internally displaced. To be considered a refugee, an individual
must have crossed an international border to seek asylum and have a “well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group
or political opinion” (Convention relating to the Status of Refugees, 1951, art. 1 § b) or have been compelled to leave his place of habitual residence to seek refuge outside his country of origin
or nationality “owing to external aggression, occupation, foreign domination or events seriously disturbing public order in either part or the whole of his country of origin or nationality.” (Convention
Governing the Specific Aspects of Refugee Problems in Africa, 1969, art. 1 § 2)
11 Yemen and Ethiopia, with respectively 192,000 and 158,000 registered refugees, are also important host countries (UNHCR, 2011: 4).
12 The presence of a family member or people from the same ethnic group, the possibility of obtaining a legal status, education and professional prospects, the proximity of the country of
origin, the existence of known routes and smuggling networks, historical and economical links and language are factors intervening in the selection of a first country of asylum by refugees, cf.
Zimmerman, 2009.
13 Between 1992 and 2000, the number of registered Somali refugees in Kenya declined from some 285,600 to 137,400 and later stabilised at approx. 150,000. The number of registered Somali
refugees in Ethiopia decreased from 514,000 in 1991 to 15,900 in 2005 (UNHCR Statistical Online Population Database). Large-scale repatriation from Ethiopian camps started in 1997. The decrease
in refugee population is not strictly linked to repatriation, as a number of refugees have also moved out of camps and informally migrated onwards and settled in Kenyan cities or in a third country.
3 Somali Displacement
18 | DURABLE SOLUTIONS
Sources: UNHCR Online Population Database, UNHCR Global Report 2011 and UNHCR’s information sharing portal on the Somali
Displacement Crisis.
Somali refugees (1987-2011)
0
200000
400000
600000
800000
1000000
1200000
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
Djibouti
Ethiopia
Kenya
Yemen
Global
DURABLE SOLUTIONS | 19
20 | DURABLE SOLUTIONS
DURABLE SOLUTIONS | 21
The large-scale UNHCR supported repatriation effort to Somaliland and Puntland from camps in eastern Ethiopia that ofcially began in
1997 highlights relevant issues for refugees, host communities, communities of return, and international aid actors that can inform
discussions of the potential repatriation and reintegration of Somali refugees.
a) Repatriation from eastern Ethiopia
In the late 1980’s and early 1990’s, the ghting and turmoil surrounding the secession of Somaliland and its rst years of independence
forced more than 600,000 Somalis to ee the northwest region of Somalia into eastern Ethiopia and Djibouti (UNHCR Statistical Online
Population Database). In the following 15 years, the majority repatriated to self-proclaimed Somaliland.
The Somali Civil War began and the rst major cracks in Siyad Barre’s regime appeared when the secessionist group the Somali National
Movement (SNM) attacked and seized control of the urban centres of Hargeisa and Burao in northwest Somalia in 1988 (Markos, 1997:
367). Supported by the Isaaq, Somaliland’s majority clan, funded by a number of Isaaq businessm people abroad, and operating out of
bases in Ethiopia, the SNM sought an independent state corresponding to the colonial borders of British Somaliland. Perceived marginali-
sation of the northwest, and the repressive and dictatorial power wielded by Barre and the Darod from Mogadishu, drove the secessionist
movement (Ambroso, 2004:5).
In 1988, having been defeated in a major battle by Eritrean separatists in the north, the head of the Ethiopian state Mengistu signed a
peace deal with Somalia’s president Barre in order to concentrate on his northern front. The peace deal included assurances that both
parties would not harbour armed rebel groups in their respective territories, depriving the SNM of its Ethiopians bases and thus provoking
it’s offensive in northwest Somalia. When Barre’s infantry was unable to retake Hargeisa and Burao, a massive bombing and artillery
campaign levelled the two cities, with an estimated death toll of 30,000 (Ambroso, 2002: 6).
Massive infrastructural destruction, violence, and state collapse caused a humanitarian crisis that displaced hundreds of thousands inter-
nally and an estimated 600,000 to the Somali regions of neighbouring Ethiopia and Djibouti (Markos, 1997: 367; UNHCR Statistical Online
Population Database). Displaced Somalis in Ethiopia were grouped in camps based primarily on their ethnicity (Markos, 1997: 375). Each
camp had a “distinct majority” from a single clan, typically situated in the territory of that clan (Ryle, 1992: 163).
When the Isaaq-led SNM was able to complete its takeover of northwest Somalia and unilaterally declare Somaliland an independent
state in 1991, large numbers of refugees in Ethiopia spontaneously repatriated, especially Isaaq clan members. Gadabursi, prominent in
Awdal region, began to repatriate in large numbers soon after, as their clan elders negotiated peace with the Isaaq leaders of the SNM
(Ambroso, 2002: 13). Spontaneous repatriation led to a signicant reduction of refugee populations in Ethiopian camps: between 1991
and 1994 some 400,000 people returned to Somaliland or dispersed locally (Ambroso, 2002: 16). The rst to repatriate were those who
had property and resources to return to in Hargeisa and other urban centres, while urban and rural poor were much more likely to remain
in the camps (Ryle, 1992: 165).
Even though refugees were returning and the actual number of refugees in camps was visibly diminishing, ofcial refugee population statis-
tics (based on the number of ration cards held in the refugee camps) remained relatively stable. In response, food rations were reduced
and a series of verication exercises were held (Ambroso, 2002: 14). According to returnees interviewed in Somaliland, a cut in rations
and services in camps in Ethiopia created a strong impetus to return (ITW with returnees, Salahley and Hargeisa, Aug. 2012).
Yet, the reduction in the refugee population proved temporary as clan-based ghting broke out in Hargeisa in November 1994 and quickly
spread to other towns. Renewed violence caused another mass displacement and forced UNHCR to stop planning for a large-scale,
assisted repatriation. Some 90,000 people ed to Ethiopia, and instability and lingering violence in Burao only subsided in 1996 (Ambroso,
2002: 7-8).
In 1997, following the stabilisation of Somaliland, UNHCR nally launched its programme to support the repatriation and reintegration
of Somali refugees from Ethiopia. Return was primarily encouraged through the provision of transportation and a return package, worth
approximately USD 130 per person and given in exchange for a ration card. The package contained food for nine months, 200 Ethiopian
birr (USD 30), blankets, jerry cans, and plastic sheeting for shelter (Ambroso, 2002: 18).
4 Repatriation efforts to Somaliland and Puntland from Ethiopia
22 | DURABLE SOLUTIONS
In addition, hundreds of Quick Impact Projects (QIPs) were undertaken in areas of return from 1994 through to 2005. QIPs focused on
providing water, healthcare, livelihoods, and education (ITW, MRRR and OCHA, Hargeisa, Aug. 2012). These projects, designed to recon-
struct and rehabilitate communities of return, generally met their short-term objectives but had little lasting impact (ITW, MRRR, Hargeisa,
Aug. 2012; Ambroso, 2002: 27) Approximately 130,000 refugees returned with UNHCR assistance between 1997 and 2005 (ITW, MRRR
and OCHA, Hargeisa, Aug. 2012).
By mid 2005, most camps had been successfully closed (ITW, OCHA and MRRR, Hargeisa, Aug. 2012). Currently, only a small number
of refugees from Somaliland remain in exile, mostly in Kebribeyah camp, near Jijiga in Ethiopia. Several refugees interviewed in the camp
explained they felt that their clan afliation could put them at risk if they were repatriating: they belonged to the Darod clan, who had been
closely allied with Barre (ITW, refugees, Kebribeyah, Aug., 2012).
Just as Somalis ed to their clan’s territory across the border, clan ties also shaped patterns of return. Many of those displaced from urban
settings chose to return to rural villages where the cost of living is cheaper; clan afliation and clan territory normally determined the choice
of a given rural area (Ambroso, 2002: 22; ITW with returnees, Salahley and Borama, Aug. 2012). People who returned to Hargeisa without
UNHCR’s support from 1991 to 1994 generally had land and resources. Those assisted by UNHCR (1997-2005) tended to be either urban
poor or rural people whose livestock had been lost, and often faced a more difcult reintegration experience (ITW with returnees, Aug. 2012).
Ofcials who worked on repatriation for the Somaliland Ministry of Rehabilitation, Reintegration and Reconstruction (MRRR) emphasised
that go-and-see visits to areas of return for refugee elders proved crucial in mobilising refugees. Elders were able to testify to their clan
“constituents” that infrastructural preparations had been made and that areas were safe (ITW, MRRR, Hargeisa and Borama, Aug. 2012).
In interviews in Somaliland with returnees, NGOs, and institutional actors, two issues emerged as critical for reintegration: land and employ-
ment. A number of returnees emphasised that they had chosen to return to rural areas rather than Hargeisa because of better access
to land. Rural returnees commonly had better reintegration experiences; they were more successful in securing land and employment.
Interviews in Hargeisa with returnees in settlements or urban poor suggested that lack of stable land tenure and employment opportuni-
ties prevented successful reintegration. Nearly all returnees stated that the toughest challenge upon return to Somaliland was lack of
livelihood opportunities and many linked this to the limited availability of land. Only a few returnees cited basic services as a major factor
in their decision making.
There was a marked difference in reported well-being and availability of opportunities between interviewees in various settlements or slums
in Hargeisa. In M. Moge camp, returnees had been allocated land, and thus been able to build permanent structures and businesses. At
State House and Stadium settlements, returnees have lived in cramped conditions in temporary structures, in some cases for more than
a decade, and lack of opportunity was widely reported. NGO informants attributed the stark contrast to difference in access to land (ITW
CCBRS, NRC, Hargeisa, Aug. 2012).
NGOs and development institutions cited livelihood and employment opportunities as the key to successful reintegration (ITW CCBRS,
NRC, MRRR, Hargeisa, Aug. 2012). Yet, unemployment is not solely an issue for returnees: Somaliland’s general population has an
unemployment rate of over 80% (ITW, MRRR, Hargeisa, Aug. 2012). Most Somalilanders were not signicantly better off than returnees
between 1997-2005, and allocating resources to programmes that beneted all Somalilanders, returnees included, might have been a
more equitable use of development resources (Ryle, 1992:166). Yet, this might not have been an option because the return package was
designed specically to induce return, as UNHCR sought closure of camps in Ethiopia (Ambroso, 2002: 15).
The speed of the repatriation exercise was an issue. As large-scale relocations began, preparations in areas of return were largely
overwhelmed. Somaliland’s MRRR informants noted that this informed the planning of the 2002-05 repatriation from Djibouti. When
repatriating refugees to the Awdal region, the MRRR and UNHCR limited the size and frequency of convoys and placed extra focus on the
development of livelihood opportunities in areas of return. These efforts appear to have paid off, as return is deemed largely sustainable in
Awdal compared with previous efforts that resulted in several large informal settlements of urban poor in Hargeisa (ITW MRRR, Hargeisa,
Aug. 2012). Thus, a fundamental tension between speed and sustainability seems to exist.
b) Observations
Somalia, with its troubled history, will require thorough assessment of areas of return to determine what programmes can aid the reinte-
gration prospects of returnees, especially in terms of employment prospects, when return becomes possible. Careful attention to issues
of land tenure and its intersection with clan territory and inter-clan relations will be essential, as will the incorporation of clan leaders in
planning and mobilisation. The historical repatriation to Somaliland also suggests that smaller and incremental population movements,
coupled with assistance and monitoring of returnees as they reintegrate, can help make return and reintegration sustainable.
DURABLE SOLUTIONS | 23
When planning repatriation to a country that is susceptible to recurrent crises, it has been suggested that “reconstruction” and “rehabilita-
tion” represent less relevant goals than “construction” and “creativity” (Ambroso, 2002: 28). Rather than framing repatriation as return to
a previous state, it might make more sense to understand a repatriation programme as assisting adaptation to a new situation. Several
informants also stressed that there were drawbacks to viewing repatriation only as a logistical exercise, rather than as part of broader
development and peacebuilding challenges (ITW MRRR, Hargeisa, Aug. 2012; Long, 2011: 30; Ryle, 1992: 167). For example, while the
repatriation was taking place, the overall economic situation in Somaliland needed to be addressed: massive unemployment was a clear
barrier for successful reintegration and required attention on a macroeconomic level.
There is also a need to ensure that repatriation is truly voluntary. Refugees should not return because life in camps has become so
challenging that they are willing to risk unsafe return or return that offers little chance of successful and sustainable reintegration (Markos,
1997: 385). Returnees who claimed that reduction of rations and services in eastern Ethiopian camps drove them to return present a
troubling challenge to the notion that their return was voluntary.
24 | DURABLE SOLUTIONS
DURABLE SOLUTIONS | 25
5 Somali refugees in Kenya and Ethiopia
In October 2011, citing the threat of Al-Shabab to regional stability, Kenya sent hundreds of troops to Somalia. Kenyan forces, still militarily
engaged there, albeit under African Union command, aimed to create a more stable border region, paving the way for possible voluntary
repatriation (BBC, 17 Oct. 2011; PANA Press, 30 March 2012). A month later, shortly after the TFG requested help from the international
community and neighbouring countries to secure Somalia, Ethiopia also crossed the Somali border to combat Al-Shabab (BBC, 19 Nov.
2011). Nearly a year later, Somalia remains unstable.
With more than half a million Somali refugees on its soil, Kenya is the main country of asylum for Somalis, followed by Yemen (219,000)
and Ethiopia (211,400). Both Kenya and Ethiopia register Somalis from south-central Somalia as prima facie refugees (Campbell, 2011:
16; Lindley, 2011: 22; RMMS). Kenya’s border with Somalia has been ofcially closed since 3 January 2007, on security grounds. While
this has not prevented refugees from seeking asylum in Kenya, it has rendered them more vulnerable (Long, 2010: 51-52).14
In both countries, a vast majority of refugees are settled in camps near the Somali border. The provision of humanitarian aid is generally
limited to the camps, where living and security conditions are challenging.
a) Overview of legal frameworks and government policies
Both Kenya and Ethiopia are party to the 1951 UN Convention Relating to the Status of Refugees, the 1967 Protocol and the 1969
Organization of African Union Convention Governing the Specic Aspects of Refugee Problems in Africa.
Article 2 (6) of the 2010 Constitution of Kenya states that any international convention or treaty will become a part of Kenyan law, thus the
provisions within these conventions are part of national law (RCK, 2012: 17-31). In 2006, the Refugees Act came into force, which shifted
responsibility for the overall management of refugees from UNHCR back to the GoK and led to the establishment of the Department of
Refugee Affairs (DRA) within the Ministry of State for Immigration and Registration of Persons (Lindley, 2011: 25). A formal registration
process and the issuance of refugee identity cards, in coordination with the National Registration Bureau, were also put in place. A new
Refugees Bill (2011) that would repeal the 2006 Refugee Act is currently being reviewed to ensure its compatibility with Kenya’s interna-
tional and regional obligations, with UNHCR’s support.
In 2004, Ethiopia adopted the National Refugee Proclamation (GoE, 2004). This proclamation put in place a legal framework and laid out
key protection principles for domestic refugee policy (UNHCR, 2011). The Administration for Refugee and Returnee Affairs (ARRA) is the
governmental branch mandated to handle refugee affairs, in close coordination with UNHCR. Some of the most signicant impediments
to refugee protection in Ethiopia are the reservations placed by the country on the Refugee Convention, which prevent refugees from
earning wages and limits their self-sufciency and ability to locally integrate (UNHCR, 2011).
b) Enhanced self-reliance and local integration
Historically, Kenya has been recognised for its generous asylum policies, allowing nearly all refugees to integrate locally. Only with the
large-scale inux of the early 1990’s did the country’s approach towards refugees become more restrictive, a relatively common occur-
rence when the size of a refugee population increases (Banki, 2004: 12, 15, 19; Campbell, 2011: 5)15.
Both Kenya and Ethiopia limit intermediate and local integration trough limiting refugees’ right to work and to move and settle freely outside
of camps either by law, de facto or through reservations placed on the 1951 Refugee Convention. This makes self-reliance and local integra-
tion virtually impossible for the majority of refugees
16
Yet, in both countries a number of refugees have settled with the local population,
outside of camps, and have found ways to integrate informally to the economy, often without approval from the government and almost
entirely outside of the legal system (Banki, 2004: 12-13; DRC, 2011: 7; RCK, 2012: 11 and 29). Refugees who live outside camps tend to
be those individuals with greater resources and education (Banki, 2004: 13 and 16). At the end of 2009, more than 160,000 unregistered
and self-supporting Somali refugees were living in Addis Ababa and other Ethiopian towns, according to ARRA (ARRA, 2009). In 2011,
UNHCR and NGOs estimated that between 80,000 and 100,000 registered and unregistered refugees, including signicant numbers of
Somalis, lived in Nairobi (Campbell, 2011: 7).
14 By closing its border, there is concern regarding the right of refugees to seek asylum.
15 Banki (2004: 19) notes that the most significant factors affecting intermediate integration or self-reliance relate to the size of the refugee population (large refugee populations are less likely
to integrate than small ones), time (the longer refugees stay in a country, the more likely they are to integrate), the political and social context, as well as individual agency. Legal, security and
economic factors (for intermediate integration) were not found to play a significant role for intermediate integration. They would, however, play a significant role for formal local integration.
16 In 2010, Ethiopia shifted its approach towards Eritrean refugees by allowing them to live outside the camps in urban locations (UN News Centre, 2010). This ‘out of camp scheme’ was open
to refugees who could support themselves financially or who had strong family networks that could provide them with the necessary assistance, since refugees were not given the right to work.
However, this policy is restricted to Eritreans and not extended to other refugees in the country (ARRA, 2010: 3; ITW UNHCR Addis, 11 June 2012).
26 | DURABLE SOLUTIONS
A number of refugees in camps complement humanitarian handouts with livelihood activities, but their income is usually not enough to
ensure self-sufciency, making even intermediate integration impossible. Activities to improve the level of refugee self-reliance have been
implemented, but major investments will be necessary for sustainability (Banki, 2004: 12; PRM, 2011; UNHCR, 2011).
In theory, in both countries, a refugee could acquire citizenship through marriage with a national (GoE, 2003: Art. 6; 2; GoK, 2011: Art. 12).
However, this is a limited avenue and the few cases put forward in Ethiopia have not been successful (ITW UNHCR Addis, 11 June 2012).
UNHCR Kenya is currently assessing potential opportunities for residency or citizenship created by the 2011 Citizenship and Immigration
Act (communication with UNHCR Kenya, Aug. 2012). For a small number of refugees with academic or vocational qualications gained in
the country of origin or asylum, work permits could eventually be secured in the country of asylum or in another country.17
c) Resettlement
From 1995 to 2010, 55,422 Somali refugees in Kenya and 2,259 in Ethiopia were resettled to a third country, most commonly the United
States (UNHCR Statistical Online Population Database). In 2011, Somalis represented nearly a sixth of the 87,849 les submitted by UNHCR
for consideration by resettlement countries (UNHCR, April 2012a: 5-6). The agency has been focussing on resettlement as a durable
solution for refugees who arrived in Kenya and Ethiopia in 1991-92 and as a protection intervention for those with specic needs (ITW
UNHCR Dadaab, Dec. 2010). Resettlement activities are ongoing in all camps, except those near Dolo Ado (ITW UNHCR Addis, 12 June
2012). It is expected that some 8,200 long-staying refugees will be resettled from Kebribeyah camp, in Ethiopia (UNHCR, 2010: 68-69).
Some 16,000 refugees in Kenya are currently at various stages of the resettlement process to the US (ITW US Embassy, June 2012).
Even though opportunities remain very limited, resettlement does bring changes into refugee camps through remittances and in terms of
options envisaged by refugees. In camps with resettlement activities, resettlement tends to become a prominent theme and a highly desired
outcome (Horst, 2006a: 152; Jansen, 2011: 13, 171). Resettlement opportunities, combined with the availability of education, might also
attract a number of refugees to the camps that are perceived as transitory spaces offering migration opportunities (Jansen, 2011: 171).
17. UNHCR statistics (11 June 2012, 3 July 2012 and 28 August 2012) indicate that 249 refugees in Dadaab have a university level education, 551 have a technical or vocational education
and 23 are listed as having a post-university level education; in Jijiga, 114 refugees are registered as having university level education, 6 post-university level, 74 technical or vocational training;
in Dolo Ado, 20 refugees have university education.
DURABLE SOLUTIONS | 27
28 | DURABLE SOLUTIONS
6 Refugees in Kenyan and Ethiopian camps
a) Overview of the camps
With the exception of Kakuma, located several hundred kilometres away from the Somali border, camps hosting Somali refugees in Kenya
and Ethiopia are located close to the Somali border. In both countries, camps are situated in isolated areas and refugees’ relationships
with host communities have been challenging. Indeed, while the opening of a camp is usually accompanied by signicant investments, job
creation and contracts for the host community, it also signies that scarce natural resources have to be shared with refugees, perceived
as beneciaries of international aid.
1. Kenya
In Kenya, most refugees in the country live in one of Dadaab’s ve camps, in the northeastern province. Kakuma, in northwestern Kenya
(Rift Valley province), is also host to a sizeable Somali refugee population (UNHCR Information portal, 11 Sept. 2012). Both camp complexes
were opened in the early 1990’s and are located in some of Kenya’s poorest regions, populated by nomadic pastoralist communities.
Dadaab’s refugee camps are situated in an area mostly inhabited by pastoralists of Somali ethnicity belonging to various Darod sub-clans,
like a vast number of the Somali refugees (Horst, 2008: 122). The northeastern province, marked by a history of marginalisation, repression
and violence, is home to 75% of Kenya’s poorest citizens (UNHCR et al., 2011: 11; Lindley, 2011: 22). When the refugee complex was
established in 1991-92, it consisted of three camps built to accommodate 90,000 refugees (Kagwanja, 2008: 221). Currently, there are
ve camps with more than 450,000 people, 98% of whom are Somalis, Dadaab has become the biggest and one of the oldest refugee
complexes in the world (UNHCR, 2010: 1 and 2012). The newest camp Ifo 2 (East and West) was opened in 2011, in response to the
large-scale inux provoked by the drought and lasting insecurity. Over 13,000 refugees are living in Kambioos.
Kakuma camp is located in northwestern Kenya, in an area populated by the Turkana that periodically requires emergency food aid, due
to recurrent droughts (Jansen, 2011: 11). The camp was founded in 1992 to receive Sudanese eeing from civil war. With the signicant
refugee repatriation to South Sudan, transfers of Somalis from Dadaab
18
and spontaneous movements of Somalis to Kakuma,
19
Somalis
have become the most populous group in the camp (Jansen, 2011: 240; RCK, 2012: 75-76). At the end of August 2012, nearly half of
Kakuma’s 101,200 residents were Somalis (UNHCR Statistics, Aug. 2012). Unlike other camps in Kenya and Ethiopia where all or nearly
all refugees are Somalis, Kakuma is characterised by its mixed population: refugees from 13 different countries, including Ethiopia, South
Sudan, the Democratic Republic of Congo and Burundi, coexist in the camp (UNHCR 2012f: 1).
Since the beginning of the year, a few hundred refugees, mostly from South Sudan, have been registered in Kakuma on a daily basis
(UNHCR, 13 Sept. 2012)
20
. The camp has now exceeded its 100,000-person capacity and negotiations for new land to enhance the
camp are ongoing with the Government.
Sources: UNHCR Statistics, June, July and Sept. 2012.
18 In 2003, a first group of approx. 12,000 Somali Bantus was transferred from Dadaab to Kakuma for resettlement purposes. In 2009, another group of some 13,000 Somalis was moved from
Dadaab to decongest the camps (RCK, 2012: 76).
19 Jansen (2011: 240) reports that from 2010, growing numbers of Somalis started arriving in Kakuma on their own from Dadaab, Nairobi and Somalia, either hoping for resettlement or trying to
avoid overcrowding in Dadaab. RCK (2012: 76) notes that in late 2011-early 2012, some 5,000 Somalis who travelled there by their own means were registered in Kakuma. This coincided with
the suspension of registration in Dadaab.
20 Registered new arrivals in 2012 include 782 Somalis, 7,965 South Sudanese, 1,606 Sudanese, 1,068 Burundians, 794 Ethiopians and 683 Congolese (DRC).
DURABLE SOLUTIONS | 29
2. Ethiopia
In Ethiopia, the government designates areas within which refugees must live (GoE, 2004: Art. 21 § 2). Until 2009, newly arriving Somali
refugees were settled near Jijiga, the capital of the Somali region, located close to the Somaliland border (Markos, 1997: 366). Jijiga
camps were rst established in 1989. Between 1997 and 2005, seven of the region’s eight camps were closed, following a large-scale
repatriation to Somaliland. The remaining refugees were moved to the only remaining camp, Kebribeyah (UNHCR, 2009). However, by
July 2007, the former camp of Aw-barre (or Teferi Ber) had to be reopened to accommodate refugees eeing an escalation of the political
crisis and the ghting between the TFG and ICU forces (IRIN, 26 July 2007). In May 2008, a third camp, Sheder, was opened. As of June
2012, Jijiga camps are home to 41,545 Somali refugees (UNHCR Information portal).
In 2009, camps near Jijiga were deemed full and new camps were progressively established in the semi-arid southern part of the Somali
region, around Dolo Ado. Since then, newly arriving refugees have been sheltered there (ITW UNHCR Addis, 11 June 2012). The number
of refugees arriving each year has steadily increased from 2009, peaking during the drought in 2011. As of September 2012, 167,482
Somali refugees live in ve camps: Bokolmayo and Melkadida founded in 2009 and 2010 respectively, and Kobe, Hiloweyn, and Bur
Amino founded in 2011 (UNHCR Information portal, 4-6 Sept. 2012). Discussions are underway between the Government of Ethiopia and
UNHCR to open a sixth camp (UNHCR, 2012k).
3. Place of origin and ethnicity
A vast majority of Somali refugees in camps are from areas relatively close to Kenya or Ethiopia. Some 92% of refugees come from one
of the ve following regions: Lower Juba (30%), Banadir (19%), Gedo (18%), Middle Juba (13%) and Bay (12%). Yet, regions of origin of
refugees signicantly vary from one camp to the other.
30 | DURABLE SOLUTIONS
In Kenya, refugees in Dadaab mostly originate from Lower Juba (42%), Banadir (Mogadishu) (21%) and Middle Juba (20%). Some 55%
of the Lower Juba refugees are from Kismayo district (more than 100,000 individuals), and more than a quarter from Afmadow district
(nearly 50,000 people). In Middle Juba, a majority of the refugees come from Buale (51% or approx. 46,000 individuals) and Sakow (32%)
(UNHCR, Sept. 2012). In Kakuma, greater numbers come from Banadir (57%). Slightly more than a quarter are from Lower Juba (26%),
and some 6% come from Middle Juba (6%) (UNHCR Statistics, June 2012).
In Ethiopia, refugees in Jijiga are primarily from Banadir (33%), Woqooyi Galbeed (11%), Hiran (10%), and Galgadud (7%). While in Dolo
Ado, refugees mostly originate from Gedo (53%), Bay (30%) and Bakool (11%) (UNHCR Statistics, July and Sept. 2012).
4. Ethnicity
More than half of Dadaab’s population belongs to one of the Darod sub-clans, especially Ogaden (more than a quarter of the camps’
population). Digil and Hawiye each represent approximately 12%, and Somali Bantu, approximately 6% (UNHCR statistics, June 2012).
Nearly 30% of Kakuma’s Somali population is Bantu. Unlike in Dadaab, Darod sub-clans represent less than a quarter of the total popula-
tion. One Somali refugee in ve in Kakuma is Hawiye (UNHCR Statistics, June 2012).
The largest ethnic group in Jijiga is Darod (40%), followed by Hawiye (24%) and Dir (8%) (UNHCR Statistics, May 2012). In Dolo Ado, the
Rahanweyn, the largest clan in the camps, represent about 58% of the Somali population. The other two primary clans are the Marehan
(22%) and Hawiye (8%) (UNHCR Statistics, Aug. 2012; UNHCR, 2012b, c, e, g, i and j).
5. Gender balance
In all camps, there are slightly more females than males aged 18-59 years. The difference is especially important in Dolo Ado where
females of this age group represent 19% of the population (31,000 individuals), while males only constitute 12% (19,000 individuals). In
Melkadida, this gender gap is even larger with 9.5% male and 19% female aged 18-59 (UNHCR Statistics, Sept. 2012). This suggests
there are a signicant number of female-headed households.
The 2010 World Food Programme, UNHCR and ARRA Joint Assessment Mission noted that in the Dolo Ado camps, some 40% to 60%
of the camps’ households were headed by females (and more than 80% in Melkadida). This could perhaps indicate that a number of
men have stayed in Somalia and a number of women in polygamous or separated marriages are registered separately as the head of the
household (WFP, 2010: 13, 17).
b) Presentation and analysis of the findings
1. Profile of households interviewed
In total, 360 households were interviewed and 410 refugees participated in 62 FGDs. Slightly more than 40% of the refugees interviewed
arrived in the camps before 2006, a third between 2006 and 2010, and a quarter after 2010. Three quarters of refugees in Ethiopia and
half of those in Kenya arrived since 2006.
This means that refugees who have arrived before 2006 are over-represented in the sample, as less
than a fth of the overall Somali refugee population in Kenyan and Ethiopian camps has been there
since 2005 or earlier. Consequently, refugees who arrived between 2006-10 and between 2011-12
are underrepresented (33% compared to 42%, and 25% compared to 41%).
A vast majority of refugees interviewed belong to one of the Darod sub-clans, especially the Ogaden,
roughly matching the ethnicity of refugees in camps. Similarly, a majority of the refugees interviewed
came from Lower Juba, Gedo or Banadir, which is also in line with the regions of origin of refugees
in camps. For most recent arrivals, Bay was also a common region of origin (more than a quarter).
1.1 Family composition: In Jijiga and Kakuma, more than 40% said they were not living with a spouse,
a much higher proportion than in Dolo Ado and Dadaab, where less than a quarter said they were
without a spouse. In the vast majority of cases, the spouse was dead or missing. Smaller numbers
were divorced. A small proportion of refugees indicated that their spouse was still in Somalia, living
in another refugee camp in Kenya or Ethiopia, or living in the US or the UK
When did you arrive in this camp?
42%
33%
25%
1988 - 2005
2006 - 2010
2011 - 2012
Year of arrival in camps ( overall populaon)
17%
42%
41% 1988-2005
2006-2010
2011-2012
DURABLE SOLUTIONS | 31
1.2 Previous experience of displacement: Slightly more than half of the refugees in Jijiga, Kakuma and Dadaab had previously experienced
displacement, while less than 10% of those interviewed in Dolo Ado had. More than half of the refugees who had arrived before 2006
said they had been displaced before. This proportion dropped to 40% for those who arrived between 2006 and 2010 and to less than a
third for the most recent arrivals.
Most refugees who had a prior experience of displacement had been internally displaced during the 1990’s, but small numbers had also
been refugees in Mombassa or Dadaab. A number of refugees in Kakuma mentioned having been in Dadaab in the early 1990’s, having
repatriated to Somalia, and been forced to ee again. In light of their previous experience, refugees recurrently stress that improvements
in stability and security would have to be long-lasting before they opt to return, in order to avoid yet another episode of displacement.
2. Livelihood activities
2.1 Livelihood prior to eeing, in exile and in the future
Prior to eeing, nearly half of the refugees interviewed in Kenya and Ethiopia were farming and/or raising livestock. Slightly more than
10% combined farming and raising livestock. Fairly signicant numbers dened themselves as merchants or petty traders. Small numbers
indicated they had a specialised trade or were artisans (masons, carpenters, mechanics, etc.).
Refugees living in camps are not ofcially allowed to work. Still, households have considerably more complex livelihood strategies than
a strict dependence on humanitarian handouts21. Aid can be understood as one of the resources at the disposal of refugees. A number
of refugees complement humanitarian assistance through engaging in various income generating activities or by nding other sources of
income notably through their social and family network. Levels of income tend to be very diverse and uctuating (Jacobsen, 2005: 28-33;
Jansen, 2011: 126, 143, 166).22
While small numbers of refugees are hired by aid organisations as incentive workers, most engage in a number of informal activities such
as trading, running businesses, providing services like catering or running restaurants and entertainment establishments. Jansen (2011:
129) stresses that such a situation is not exceptional in Sub-Saharan Africa where the informal sector accounts for an average of 78% of
non-agricultural employment.
Approximately 45% of households interviewed have at least one liveli-
hood activity in the camp, with slightly more than one household in ten
having more than one source of income. In a very small proportion of
households, parents reported that some of their children were working.
Such a proportion of households with at least one livelihood activity
is comparable to gures found in Dadaab camps by the American
Bureau of Population, Refugees, and Migration during a recent liveli-
hood assessment (PRM, 2011). Some 43% of households interviewed
by PRM earned money through economic activities.
Based on household interviews and FGDs, refugees who have been in exile for longer appear more likely to have a livelihood activity and
more established refugee camps offer more livelihood opportunities.
In the most recently opened camps in Dolo Ado, opportunities are
scarce. Slightly more than half of households arrived before 2006 and
nearly half of those arrived between 2006 and 2010 reported having
at least one livelihood activity. Less than 30% of refugees arrived in
2011-12 said so. In all cases, women were less likely to work then
men. Based on FGDs, adults under 35 seemed more likely to have a
livelihood activity than older people.
21 Jansen (2011: 143) suggests the following twelve livelihood categories: businesspeople, incentive workers, employees in the informal sector, remittance receivers, shifters (people who secure
an income or other opportunities from another city and partly shift between their homelands, other locations in Kenya and the camp), the poor (refugees whose only source of income is aid
handouts), entrepreneurs (craftsmen and women who run small home-based shops or prepare food for road-side shops, or are engaged in weaving, cloth making, stamp making, etc.), refugee
“elites” (people able to obtain assets and cash from other sources due to their status or position of power in the camp), hustlers, crooks and petty thieves, the dependants, the targeted (people
able to make use of schemes, funding and other arrangements designated specifically for them on the basis of vulnerability or another social label, those who become eligible for micro-credit
after completing vocational training programs, the insecure who are given special protection arrangements, or those who are eligible for scholarships), the unrecognised.
22 Accurate data on livelihoods tends to be relatively difficult to obtain: refugees might fear that disclosing their level of wealth will affect the support of agencies or put them at risk; a number
of income generating activities are not legal or taboo, disclosing corrupt practices might endanger the recipient (Jansen, 2011: 130).
HH income in exile
0%
20%
40%
60%
80%
100%
1988- 2005 2006- 10 2011- 12
No income
At least one sorce of
income
Proporon of men and women with a livelihood
0%
10%
20%
30%
40%
50%
1988 - 2005 2006 - 2010 2011 - 12
Men
Women
32 | DURABLE SOLUTIONS
While the most common livelihood activities identied by PRM in Dadaab were raising livestock (27% of active heads of household) and
farming (20%), only a small proportion of the refugees interviewed mentioned such a source of income. This could be explained by restric-
tions placed on such activities either by law or through a lack of access to arable land or pasture in most refugee settings studied. For
example, refugees in Kakuma are not permitted to own livestock or take part in farming activities (Jansen, 2011: 126). While in principle
this policy also holds true for Dadaab, it is not being enforced (UNHCR, written communication, Aug. 2012).
Of the households interviewed who had at least one livelihood activity, a small proportion (less than 7 %) said they were farming or raising
livestock. The most common activities were petty trade (37% of households with a livelihood activity). Slightly more than a fth earned
money through a specialised trade, such as carpentry, tailoring, masonry or hairdressing. Some 19% of households earned money as
incentive workers or teachers, slightly fewer as shop owners and traders (17%) or casual labourers (15%).
In FGDs, the most common livelihood activities mentioned by refugees were operating small shops (grocery stores, tea shops, butcheries),
doing petty trade (milk selling, recycling of plastic bags, milk selling, etc.) or working as casual labourers. Smaller proportions were hired
as teachers and incentive workers. In the vast majority of cases, refugees reported receiving cash for their work. Small numbers were
receiving a combination of cash and goods, or only goods.
A minority of refugees mentioned receiving support for their livelihood activities, mostly from humanitarian organisations. A few individuals
also cited support from relatives, friends or the host community. This information could not be assessed against statistical data, as such
information does not seem to be available. However, it does appear that the proportion of refugees receiving support for their livelihood activi-
ties is relatively small. For example, recent mapping of livelihood activities in Jijiga camps shows that less than 3,000 refugees benet from
any kind of livelihood-related support, which represents less than 7% of the camp population (data compiled by DRC Jijiga, 24 Sept. 2012).
2.1.1 Livelihood in the future
If refugees were to return or move out of the camps, they would mainly plan to farm or pursue business activities. Others would engage
in petty trade, raise livestock or use their skills in various elds such as carpentry, tailoring, masonry or hairdressing. A greater propor-
tion of women plan to farm, while more men intend to raise livestock. The relatively low numbers of people planning to raise livestock
can probably be partly understood by the fact that refugees have often lost their livestock to the drought or when eeing. Parents most
frequently mentioned that their children would pursue business activities or petty trade, farm or raise livestock.
Total respondents: Female: 157; Male: 181
In FGDs, refugees indicated that they plan to pursue business activities, farming or raise livestock. Those planning to run business stressed
the need for grants, while those who aim to farm or raise livestock highlighted they need money, but also agricultural inputs, animals and
land. Refugees’ intended livelihood activity would affect their needs in terms of access to land.
2.2 Remittances
A number of refugees benet from assistance from friends or relatives living outside the camps. Cindy Horst assessed that at least 10 to
15% of refugees in Dadaab receive remittances. More than a third (37%) of the refugees interviewed by PRM mentioned receiving such
support (PRM, 2011: 3, 28-29; Horst, 2006: 29).
Less than a fth of households interviewed for this research said they were receiving remittances, generally in the form of cash, but in a
minority of cases in the form of food. Remittances are usually sent on a monthly basis by relatives in the US or Kenya, but also in a number
of other Western countries (and Somalia, in a few instances).
Acvity planned for the future
0
50
100
150
200
250
Farming Raising
livestock
Shop/
business
Pey trade Trades Other
Men
Women
DURABLE SOLUTIONS | 33
Greater numbers of refugees in Kenya mentioned receiving remittances, ranging from USD 45 to 3,000, than in Ethiopia (26% against
10%). This difference seems related to time in exile, rather than location: refugees who have been living in the camp for longer appeared
more likely to receive remittances. More than a quarter of those who arrived before 2006 said they were receiving support from relatives
outside the camp, less than a fth of those who arrived between 2006 and 2010 less than a tenth of those who arrived in 2011-12. A
large number of refugees in Ethiopia have only arrived in the camps recently.
Of all people interviewed, only 14 individuals, nearly all in Kenya, indicated sending support to their family, mostly in Somalia, but also in
Kenya and Ethiopia. Amounts sent range from USD 15 to 350 a year.
In FGDs, a small number of refugees mentioned receiving remittances and even smaller numbers said they were sending money to relatives
in Somalia.
3. Education and Vocational training
3.1 Education
In all camps, a vast majority of refugees have never accessed education
either before eeing or once in exile. For example, more than 75% of
refugees registered in Dadaab, 67% of refugees in Jijiga and 93% of
those in Dolo Ado reported having no education (UNHCR statistics,
June, July and Aug. 2012).
In Ethiopia, less than half of the refugees interviewed indicated that
one or many of their family members had attended school in the
camps, compared with 64% in Kenya. Figures were especially low
in Dolo Ado (41% of refugees), which could be explained by the fact
that schools have only opened recently in some of the camps, only emergency ones in Buramino. In July 2011, many months after the
drought-related emergency started, up to 80% of the Dolo Ado camp children were out of school because there were still no schools in
several camps (UNHCR, 2011a
The proportion of households reporting that at least one family member has attended school in the camp increases with the duration of
exile. More than two thirds of those who have arrived before 2006 had at least one family member who had been to school, a proportion
that dropped to slightly more than half for those who arrived between 2006-2010 and less than half for those arrived in 2011-12.
Such gures are not unexpected. Even though education facilities have been set up in all camps in Kenya and Ethiopia and in both countries,
pupils follow the national curriculum and sit for national exams, for school-aged children, access to quality education is unsatisfactory
(UNHCR et al., 2011: 7; ITW DRC Jijiga, Sept. 2012). Education tends to be underfunded and educational facilities are insufcient to meet
the needs of the refugee population. Those that exist are in poor condition and under-equipped, and teachers are under-qualied. Enrol-
ment, attendance and levels of retention are low, especially for girls (Dippo, 2011: 19; UNHCR et al., 2012: 9, 15, 25; UNHCR, 2011a: 31)
23
.
In Dadaab, where efforts are being made to ensure that educational activities prepare refugees for life outside the camps through a recently
adopted joint education strategy, 39.7% of school-aged children were enrolled in primary school and 8.6% in secondary school in June
2012 (UNICEF/UNHCR, 2012: 2 and 7).24 In Kakuma, 68% of children are attending primary or secondary school (UNHCR et al., 2012:
6,13). In Jijiga in 2010, less than 45% of school-aged children were enrolled in primary school. For secondary education, gures dropped
to 17% of girls and 31% of boys (UNHCR, 2011a: 30).
In FGDs, refugees frequently mentioned that parents who had the means to do so were sending their children to private schools in the
camps. In Dadaab, in 2009, 3% of the school-aged population were enrolled in such schools (Dryden-Peterson, 2011: 60). Refugees
participating in FGDs explained that the quality of education is better in private facilities, as teachers are better trained and classes are
smaller.25 Other reasons cited by parents for sending their children to private schools in Dadaab in 2009 were strong discipline and the
integration of religious and secular studies (Dryden-Peterson, 2011: 60).
23 Dippo et al. (2012: 10-11) note that barriers confronting girls’ access to education include social-cultural norms that place girls in a subordinate position to boys, threats to security and lack
of facilities that accommodate the health needs of girls and the fact that teachers in the camps are mainly men, contributing to cultural barriers of communication and a shortage of female
role models.
24 In a recent assessment in Dadaab, NRC (2012: 4) noted that low attendance in public schools can notably be explained by the fact that a number of students prefer attending religious
schools, a lack of resources for purchasing school uniforms and learning material, long distances to school, language barriers and competing priorities.
25 Ratios of teacher to students in public schools in camps are often very high (1:131 in Dadaab against a standard ration of 1:40; up to 1:90 in Kakuma (UNICEF/UNHCR, 2012: 4; UNHCR et al.,
2012: 9)). The number of trained teachers in the camps’ school system is too low to meet the needs. In Dadaab and Kakuma, less than 20% of teachers have had formal training and numerous
teachers stop teaching when they secure a better paying job or a less demanding one in another sector (NRC 2012: 1; Dryden-Peterson, 2011: 56; UNHCR et al., 2012: 24).
Proporon HH with at least one member having
aended school in the camp
0
10
20
30
40
50
60
70
80
1!
Daadab!
Kakuma
!
Jijiga!
Dolo Ado!
34 | DURABLE SOLUTIONS
Only a minority of children graduating from primary school have access to secondary education, a gap regularly underlined in FGDs.
According to Cooper (2005: 465-66), in Dadaab in 2003, only 27% of those who completed primary school could pursue further
education.26 At that point, it was estimated that by 2009, more than 11,000 refugee youths in the camps would have completed primary
school without the possibility of enrolling in post-primary education. In Kakuma, only 100 to 200 secondary school positions are available
every year. As the number of primary school graduates exceeds the available spaces in secondary schools, numerous students are unable
to continue their education (UNHCR et al., 2012: 21). Even though higher education for refugees is deemed important both for individuals
and for fostering leadership in protracted settings and post-conict reconstruction, opportunities for post-secondary learning are even
more limited (Dryden Perterson, 2011: 52).27
3.2 Vocational training and adult education
Vocational training classes are offered in most camps. For example, in Dadaab, the Norwegian Refugee Council offers a one-year full
time Youth Education Pack (YEP), focused on literacy, life skills and vocational training (ITW NRC Dadaab, 4 June 2012). In Kakuma, Don
Bosco provides trainings to some 600 students a year in carpentry, masonry, welding, plumbing, computers, electrical work, agriculture,
tailoring and dressmaking, as well as secretarial and typing courses (Don Bosco, 2011: 6 ). In Jijiga, a handful of organisations offer basic
training and support to livelihood activities.
The proportion of interviewed refugees who mentioned having attended vocational training classes in camps was relatively low. Nearly
twice as many refugees in Kenya than in Ethiopia mentioned having attended vocational training classes (28% against 15%). The greatest
proportion of people who reported having attended vocational training was in Kakuma camp, with 36% of refugees interviewed, and the
smallest, in Dolo Ado. This can be explained by the fact that while vocational training activities are limited in all camps and deemed insuf-
cient by refugees met in FGDs, they are especially so in Dolo Ado’s recently opened camps.
In general, just as with education, the proportion of refugees who have attended vocational training classes increases with the duration
of exile (29% for 1989-2005; 23% for 2006-2010; 12% for 2011-12). Based on FGDs, refugee leaders, entrepreneurs and employees
seem to have slightly better access to vocational training, while opportunities for women and vulnerable refugees appear extremely limited.
Refugees interviewed had most commonly attended business management and tailoring trainings. Small numbers also participated in
computer, counselling, teaching, carpentry of mechanical training. Some 75% of those who beneted from vocational training classes
assessed that they had helped them improve their lives or increase their earning due to the development of new skills and business and
job opportunities.
In FGDs, refugees listed a number of trainings that would be benecial, most commonly tailoring, business training, carpentry, driving
courses, computer courses, electrician and mechanics training. Language classes were also regularly requested. It was stressed that
refugees hired as incentive workers could benet from more advanced on-the-job training, skills transfer and mentoring, leading to
enhanced responsibilities.
A majority of refugees registered in camps have had no access to education before eeing (UNHCR statistics, June, July and Aug. 2012).
In FGDs, refugees regularly highlighted that adult education and literacy should be more readily available in the camps. Adult education
centers exist in some camps, but are deemed insucient by refugees. For example, in Kakuma, 12 centres run by the refugee community
offer language classes and a number of courses in topics such as mathematics, science or business (UNHCR et al., 2012: 39).
3.3 Skills and social relations gained in exile
Only a third of the refugees mentioned gaining new skills in exile that would be useful if they were repatriating or moving out of the camps.
Skills most commonly mentioned in interviews and FGDs were education, business skills and tailoring. Small numbers reported gaining
carpentry, construction and brick-making skills, computer education, life-skills, leadership and peacebuilding skills or watchmen skills. A
number of refugees in Kakuma underlined having learned English and/or Swahili.
While nearly half of the refugees in Kenya mentioned having gained useful skills in exile, only 13% of refugees in Ethiopia felt the same way.
In general, refugees who had been there for longer were more likely to indicate they had gained useful skills in exile (44% of those arrived
before 2006; 32% of the 2006-2010 group; and 16% of refugees arrived since 2011).
A number of refugees indicated having gained valuable social relationships in exile. Refugees notably mentioned having developed business
relations, having been part of workers and traders associations and having interacted with people of other nationalities.
26 The Convention on the Rights of the Child provides that only primary education is mandatory (art. 28 § a).
27 For example, Afghan refugees who had had access to higher education repatriated earlier, took up work as civil servants or as NGO managers, filling critical roles in a society during a
rebuilding process (Dryden Perterson, 2011: 52).
DURABLE SOLUTIONS | 35
4. Safety
Only a minor proportion of refugees interviewed in Ethiopia (3% in Jijiga
and 20% in Dolo Ado) felt it was unsafe for them or their family to travel
outside the camps. This proportion was greater in both Kenyan camp
locations (57% in Dadaab and 65% in Kakuma).
In Kenyan camps, refugees commonly explained that it was unsafe to
travel outside the camps because they did not have the authorisation to
do so. In Dadaab, a number of refugees also stressed they feared police
harassment. Refugees in the other camps surveyed did not express
such concern.
5. Intentions and pre-conditions for repatriation
5.1 Reasons for displacement
Nearly all refugees mentioned ghting as one of the reasons that led to their displacement (87%). Insecurity was raised by more than 40%
of refugees, while small numbers also cited drought, loss of livestock and loss of livelihood.
Refugees who arrived in 2011-12 were more likely than others to list drought and/or loss of livestock as one of the reasons leading to
their displacement. Greater proportions of refugees that arrived before 2011 indicated that ghting had caused their displacement. Nearly
90% assessed that the reasons that led to their displacement still prevailed. A slightly higher proportion of those who arrived in 2011-12
compared to refugees who arrived before said so.
5.2 Intentions for the future
When asked about their intentions for the future, nearly half of those interviewed wanted to move to another country (49%), 31% said they
would return to Somalia and 20% that they would like to stay in Kenya or Ethiopia. Such proportions match the intentions expressed in FGDs.
A relatively small proportion of recent arrivals (2011-2012) said resettlement was their plan for
the future. A vast majority wanted to either return to Somalia or integrate locally. The proportion
of people hoping to move to a third country increased substantially with time, while the desire to
integrate locally decreased. Less than a fth of refugees who arrived in the camps before 2005
mentioned intending to repatriate.
In the recently opened Dolo Ado camps, the only camps with no resettlement activities, 47% of
refugees said they would like to return to Somalia, while 42% want to settle permanently in Ethiopia.
Only 11% intended to move to a third country. In Jijiga, this picture is completely different: 80%
said they wanted resettlement, 17% wanted to repatriate and 3% wanted to stay in Ethiopia. In
Dadaab and Kakuma, slightly more than half of the refugees interviewed hoped for resettlement.
Refugees feeling unsafe to travel outside the camps (%)
0
10
20
30
40
50
60
70
Daadab Kakuma Dolo Ado Jijiga
Why were you inially displaced
0
10
20
30
40
50
60
70
80
90
100
Fighng Insecurity Drought Loss of
livestock
Loss of
livelihood
Propoons of respondents
1988-2005
2006-2010
2011- 2012
Why were you inially displaced
312
157
82
50 47
0
50
100
150
200
250
300
350
Fighng Insecurity Drought Loss of livestock Loss of
livelihood
Total respondents: 358
What are your intenons for the future
20%
31%
49%
Becoming a resident in this country
Going back to Somalia voluntarily
Reseling somewhere else
36 | DURABLE SOLUTIONS
Proporon of refugees who would not consider returning
0
10
20
30
40
50
60
70
80
Dadaab
Kakuma
Jijiga
Dolo Ado
Figure: Intentions for the future by camp and by period of arrival
In FGDs, youths and older arrivals were the least likely to express the desire to repatriate and more commonly said they wanted to move
to a third country. Refugee leaders and the most recent arrivals (2011-12) were most likely to want to repatriate.
When asked if they would consider returning to Somalia if conditions were conducive, 53% of households interviewed answered positively.
The longer refugees have been in the camp, the less likely they are to consider return. Less than half of households who arrived in the
camps before 2005 said they would consider returning if conditions were conducive (45%), while more than half (55%) of those who have
arrived since 2006 said they would. A slightly greater proportion of refugees who would consider returning listed drought, livestock and
loss of livelihood as part of the reasons for their displacement.
While Jijiga had the highest proportion of refugees assessing that reasons that led to their displacement no longer prevailed (27%), this
did not inuence people’s intention for the future: Jijiga also had one of the highest proportion of people saying they would not consider
returning, even if conditions were favourable (68%). This could be partly explained by the large-scale resettlement of long-staying refugees
currently taking place in one of the Jijiga camps, Kebribeyah (UNHCR, 2010: 68-69).
Unlike the duration of exile, having been displaced before, receiving remittances or having beneted from education or vocational training
classes, did not seem to inuence intentions. While belonging to a
minority did not appear as a signicant factor in household interviews,
it was regularly mentioned as an impediment to return in FGDs.
A majority of the refugees who said they would not consider returning,
indicated they were waiting for resettlement, while more than a third
wanted to become residents in their country of asylum. A number
of refugees assessed that life in exile was better than they had in
Somalia: in camps they have access to school and health facilities
and have managed to nd some livelihood opportunities. Others
said they had traumatic experiences in Somalia and would never
consider returning.
5.3 Pre-conditions for return
Refugees commonly stressed that improvements in stability and security would have to be long-lasting before they consider returning.
This can notably be explained by the fact that a number
of refugees in camps have experienced several episodes
of displacement, due to short-lived improvements in
security and stability, and will do their utmost to avoid
being displaced yet again. Refugees’ decision to repat-
riate or not is also likely to rest on several other factors,
including access to education and health care, livelihood
opportunities and whether humanitarian assistance would
be provided upon return.
If yes, what would be the pre-condions for taking the decision to repatriate
152
61 47 55
25 24
44
14 314 27
Security and safety
Access to educaon
Access to healthcare
Livelihood opportunity
Restoraon of land
Restoraon of housing
Humanitarian assistance
Assistance from relaves
Assistance from community
Access to jusce
Access to food and water
Proporon by period of arrival
0
20
40
60
80
Repatriate Integrate
locally
Resele to a
third country
Repatriate Integrate
locally
Resele to a
third country
1989-2005
2006-2010
2011-2012
Proporon by camp
0
20
40
60
80
100
Dadaab
Kakuma
Jijiga
Dolo Ado
DURABLE SOLUTIONS | 37
5.4 Location of return
Only a small number of refugees said that if they were to repatriate, they would return to a different location than their locality of origin.
Those who said so typically came from a rural area, but planned to settle in an urban area for livelihood opportunities, the urban way of
life, security and access to services. The presence of relatives, access to housing and land were only underlined in a few instances. The
main motivations for returning to the locality of origin were the presence of relatives and friends, owning land, housing or other property
and livelihood opportunities.
Reasons why people would return to their locality of origin varied slightly between Kenya and Ethiopia. In Kenya, refugees mostly cited the
presence of relatives, security and livelihood opportunities. Refugees in Ethiopia most often highlighted having access to land in the locality.
5.5 Land and other assets in Somalia
Approximately half of the refugees interviewed had access to
land in Somalia before eeing. Of that number, slightly less than
half said they would still have access to it, a fth did not know. In
nearly all cases, when refugees said they would still have access
to their land, they also indicated that a relative was looking after
it. Approximately a third of the refugees in FGDs believed they
would still have access to land in Somalia. Some 40% said a
relative was looking after it and they would still have access to it.
Three quarters of the refugees in Dolo Ado indicated they had access to land before eeing. Some 64% believed they would still have
access to it, nearly a fth did not know, and another fth said they would no longer have access to it. Figures in Jijiga were very different:
only a quarter of refugees said they had access to land before eeing. Of those, 5% indicated they would still have access to it and 21%
said they did not know. In Kakuma and Dadaab, 48% and 57% of refugees said they had access to land before eeing, 40% in Kakuma
stated they still had access to the land while in Dadaab, 57% indicated they would still have access to it.
Differences in gures seem to be partly linked to the
duration of exile. The proportion of refugees who had
been in camps for longer (1988-2005) who said they
did not have access to land in Somalia is greater, as is
the proportion of those saying that they would not have
access to it. This also surfaced in FGDs, along with
the fact that youth and refugees belonging to minority
groups are less likely than other refugees to still own
land in Somalia.
Less than a quarter of the refugees interviewed said they
still owned other assets in Somalia, most commonly
a house or a farm. Small numbers also had livestock.
Virtually all refugees in Jijiga said they had no other
assets left in Somalia, while less than a quarter did in
Daadaab and Kakuma. In Dolo Ado, 40% did, most in
the form of a farm. Just as with land ownership, refugees
who recently arrived (2011-2012) were more likely to
still own assets in Somalia (44%, compared to 19% of
those arrived between 2006-2010 and 13% of those
arrived before 2006).
A slightly greater proportion of refugees considering returning had access to
land (58% compared to 50%) or other assets (29% compared to 15%) and
would still have access to it (50% compared to 43%).
5.5 Assets in the camps
Only a very small proportion of refugees (8%) reported having brought assets
with them when eeing from Somalia. Several refugees said they had no assets
in the camps that they would bring with them if they were leaving. Those who
would commonly listed non-food items, money and shelter. Over a fth of those
who would consider repatriating, would plan to leave with their shelter.
Dadaab
Kakuma
Jijiga
Dolo Ado
Propoon of refugees who had access to land before fleeing
0
10
20
30
40
50
60
70
80
Would you sll have access to it?
0
10
20
30
40
50
60
Proporon
Did you have access to land before fleeing?
0
20
40
60
80
100
Yes No Does not know
Proporon
1988-2005
2006-2010
2011-2012
Yes No Does not know
1988-2005
2006-2010
2011-2012
Assets that refugees would take with them if they were repatriang
or moving out of the camp
28
17
119
68 64
0
20
40
60
80
100
120
140
Livestock Farming
tools
NFIs Money Shelter
Total respondents: 287
38 | DURABLE SOLUTIONS
Refugees without relaves in Somalia (%)
0
5
10
15
20
25
30
35
40
1989-2005
2006-2010
2011- 2012
5.6 Support that could be provided by humanitarian organisations
When asked how humanitarian organisations could help refugees prepare to repatriate and reintegrate if conditions became conducive
in Somalia, heads of households mostly focused on direct support to repatriation and reintegration, commonly listing the provision of
transport, food, shelter and cash.
In FGDs, refugees suggested that aid organisations should take measures in the camps to help them be ready to repatriate and reinte-
grate, notably by ensuring a greater access to education and vocational training, as well as providing livelihood opportunities. Incentive
workers stressed that they could further develop their skills and capacities through mentoring by aid organisations, which would enable
them to nd employment upon return.
Refugees also highlighted that humanitarian organisations should be ready to support repatriation and reintegration (through transport
and the provision of basic assistance) and assist refugees with access to land and housing upon return. In addition, refugees commonly
stressed that improvement of peace, justice and stability would be crucial to help them repatriate and reintegrate.
6. Decision-making and sources of information
6.1 Decision-making process
In all camps, the decision to repatriate or not would be taken by heads of households. To a limited extent, relatives in Somalia, refugee
leaders and elders would also be inuential in such a decision.
Refugee leaders were mentioned as inuential in decision-making more regularly in Jijiga, while relatives in Somalia were more signicant
in Dolo Ado. It was a regular occurrence for recent arrivals (2011-12) to indicate that they would also take advice from relatives in Somalia
into account.
6.2 Information
Some 70% of refugees said they felt well informed about the situation in their area of origin. Similar proportions were recorded in FGDs,
although women and vulnerable refugees felt well-informed in remarkably smaller proportions, while youth and entrepreneurs were the
most likely to feel well-informed. The main source of information was media (mostly the BBC, VOA, Universal TV and Star FM), followed
by relatives in the locality of origin and other refugees. Smaller numbers listed telephones, the Internet (and social networks) and new
arrivals as sources of information.
Nearly half of the refugees interviewed indicated receiving information from their area of origin daily, while 13% said they never received any
information. In Dolo Ado, a quarter of the refugees said they never received any information, which could partly be explained by the fact
that in a number of the camps, there is no phone network, which considerably limits contact with relatives in Somalia. Yet, in all camps,
media are the most common source of information. Being well informed or not did not seem inuence return intentions.
6.3 Relatives in Somalia
In all locations except Jijiga, only a minority reported no longer having relatives in Somalia (between 3% and 22%). In Jijiga, more than half
of the refugees said so. Some 70% of the refugees who participated in FGDs said they still had relatives in Somalia.
The proportion of people who said having no known relatives in Somalia
increased with the duration of exile. While only 11% of those who arrived
between 2011-12 said they had no relatives in Somalia, more than a third of
the older arrivals said they did have. This was also noted in FGDs. Some 75%
of those saying they would consider returning to Somalia still had relatives in
the country, compared with 68% of those who would not consider returning.
Temporary visits and spontaneous repatriation to Somalia
Only a very small number of refugees who were interviewed or participated in
FGDs reported having been back to Somalia since eeing (6% of the population interviewed and less than 3% of the FGD participants).28
Those who said they had visited did so for family-related reasons or to look after their land or assets. Most reported having been only
once, but a few mentioned returning once a year. In slightly more than half of the cases, people who had visited said the situation had been
28 Eliciting answers to such a question can be problematic as refugees are usually aware that visits to their country of origin can be seen as in contradiction with their claim for protection in
another country.
DURABLE SOLUTIONS | 39
characterised by ghting and insecurity, while a small number said that at the time of their trip, it had been calm or the situation had been
improving. Problems most frequently experienced during visits related to ghting and insecurity, the poor quality of roads and transport
and insufcient access to food and water.
Refugees who had never been back to Somalia mostly indicated that insecurity and ghting had prevented from doing so. Yet, in FGDs,
the lack of means to pay for transport, the absence of transportation, the fact that assets have been looted and lands occupied and bad
memories were also commonly raised.
Very few people knew refugees who had repatriated since they had arrived in the camps (less than 30 individuals). Similar proportions came
out of FGDs. Those who knew people who repatriated mostly mentioned harsh conditions, insecurity and lack of livelihood opportunities
in the camps, the improvement of security in Somalia, family bonds in Somalia and the will to resume agricultural activities as reasons for
their return. In FGDs, youth also stressed that other youth had repatriated to take advantage of job opportunities in Somalia.
7. Observations
While refugees might share a number of religious and cultural practices, as well as a collective history and memory, their experience will
be gendered and refugees are likely to be different in terms of personal history, skills and interest, socio-economical status, health, and
social belonging or beliefs (Colson, 1999: 23; Essed, 2004: 2). The uniqueness of refugees’ experience means that no response or analysis
can be appropriate for all.
When examining return intentions and the bonds Somali refugees have with their country of origin, the length of their exile, drivers of
displacement, conditions in the country of asylum, age, and belonging to a minority group are important elements.
Given the choice, nearly half of refugees would move to a third country. Yet, if conditions were conducive in Somalia, a signicant propor-
tion would consider returning. Local integration does not emerge as a popular option, possibly because refugees deem their conditions
in asylum challenging. However, intentions and reintegration prospects seem to signicantly vary with the duration of exile, just as bonds
with Somalia do.
Unsurprisingly, refugees who have been in exile for longer seem to have a weaker bond with their country of origin and might have limited
reintegration prospects. While more than half of refugees in prolonged exile said they still had relatives in Somalia, only a minority believed
they would still have access to land or other assets.
In general, adults below 35 and refugees belonging to minority groups seemed less likely than other refugees to still own land or other
assets in Somalia, which could partly explain why these groups expressed little desire to repatriate, even if conditions were conducive.
The fact that many youths have spent most of their lives in exile and have limited recollection of Somalia, if any, must also come into play.
For youths who grew up in camps, which can to some extent be compared to cities (cf. Jansen, 2011), returning to the rural areas that
many come from, with limited public facilities, seems especially difcult to envisage. Further research would be required focused on youth
which was beyond the scope of this study.
Refugees who ed after having experienced direct violence appeared especially unlikely to consider repatriation, regardless of the duration
of their exile or their bond with Somalia. Such views were repeatedly expressed by refugees in Bokolmayo, the rst camp opened in the
area of Dolo Ado that hosts a signicant number of refugees who ed from violence in Banadir in 2009.
The location of exile did not seem to signicantly inuence refugees’ perspective. Yet, in all locations, refugees stressed that conditions
were difcult and could induce premature returns. In Dolo Ado camps, a number of refugees stressed that others had left due to the
harshness of the setting.
Having been displaced before, receiving remittances or having beneted from education or vocational training classes did not seem to
signicantly inuence intentions.
Refugees usually note that the causes of their displacement still prevail. Measures to enable refugees to reintegrate in Somalia or integrate
elsewhere could already be taken in the camps by increasing access to education and livelihood opportunities.
40 | DURABLE SOLUTIONS
DURABLE SOLUTIONS | 41
7 Perspectives of selected communities of return
The situation in Somalia broadly differs from one location to another. Whereas Somaliland and Puntland are relatively stable and
conict-free, most of the south and central region remains subject to instability, violence and continued population displacement. A majority
of Somali refugees in Kenyan and Ethiopian camps ed from the southern part of Somalia: most come from Lower Juba, Banadir, Gedo,
Middle Juba and Bay.
While the security landscape in south-central Somalia has been changing rapidly since the beginning of 2012, Human Rights Watch recently
stressed that there is no evidence of a signicant improvement in the human rights situation for the local population (HRW, July 2012).
OCHA noted that increased military operations in late 2011 and the rst half of 2012 resulted in localised and temporary displacement in
Gedo, Lower Juba, Bay, Bakool and Hiran regions. Yet, the organisation reported an improved humanitarian access in Mogadishu, Gedo
and Dhobley in Lower Juba (OCHA, July 2012: 1-2).
After two decades of conict and instability and in the absence of a functioning government, public infrastructure and basic social services
are generally very limited and in a poor state in south-central Somalia. Millions of Somalis have no access to services such as WASH,
health or education (OCHA, July 2012: 22).
In order to capture key elements to consider when reecting on the eventual repatriation of Somali refugees, Somali partners (Sean
Deveruxe Human Rights Organisation, Somali Children Welfare and Rights Watch and Gedo Social Development Organisation in Belet
Xaavo) conducted 73 interviews with key informants in selected districts. Key informants included community leaders, elders, clan leaders,
local authorities, service providers (health and education), entrepreneurs and civil society organisations.
The three districts, Afmadow in Lower Juba, Baidoa in Bay and Belet Xaawo in Gedo, were selected based on access and the areas of origin
of refugees in camps. Although experience has shown that refugees may not be repatriating to the locality they came from, it is likely that
a number would be returning to their region of origin if the situation allows, especially if they still have assets or relatives (Long, 2010: 5-6).
The following section sketches the situation in selected communities of return and introduces the perspectives of key informants on the
potential large-scale return of refugees and IDPs. It highlights that access to land and skills building for returnees, allowing them to secure
a livelihood and contribute to the reconstruction of their communities, are important aspects to consider when discussing repatriation
and reintegration.
a) Overview of the situation in selected districts of return
1. Afmadow in Lower Juba Region
1.1 Security and stability and humanitarian access
In May 2012, Somalia and AU forces captured the town of Afmadow, home to some 50,000 people mostly from the Ogaden sub-clan
(Darod) in Lower Juba (est. pop.: 386,000), from Al-Shabab (BBC, 31 May 2012; UNDP, 2005). The group had taken the city from Hizbul
Islam in November 2009 (IRIN, 24 Nov. 2009). During the rst six months of 2012, insecurity of mid to high level intensity, notably stemming
from inter-clan ghting, was widespread in the district (FSNAU, 2012; UNDP, 2005). In interviews, key informants described the security
in Afmadow as average, but improving. They partly attributed the responsibility for the remaining insecurity to military forces present in the
area. Local authorities reported the presence of mines or unexploded ordnance on roads surrounding the town.
In 2006, after Afmadow was taken over by the Islamic courts, many humanitarian organisations left and have not resumed activities.
However, organisations that had been operating in the region the longest have been allowed to remain (Sedhuro, 2012). Even though OCHA
reports that humanitarian access has been improving, it is still considered extremely restricted (OCHA, July 2012: 2; OCHA, Aug. 2012)
1.2 Displacement
It is estimated that some 31,000 people are displaced in Lower Juba (OCHA, July 2012: 22). IDP gures for Afmadow district provided
by local authorities, civil society organisations and clan leaders vary between 4,500 and 15,000 people living in three settlements or with
the host community.
Some 200,000 people from Lower Juba have become refugees, mostly in Kenya. Some 41% of the population in Dadaab camps originate
from Lower Juba. Slightly more than a quarter are from Afmadow (approx. 48,000 people) (UNHCR, Sept. 2012). Several key informants
mentioned that the population had become scarce in the last two decades. They usually listed several periods of displacement: during
42 | DURABLE SOLUTIONS
Community Driven Recovery, Development and Safety programming in Somalia
As an example of how a local governance programme can consider returns, DRC and Danish Demining
Group have been implementing community driven recovery, development safety programmes in
Somalia since 2008. Through the community-driven process, communities analyse their resources
and needs, prioritise their requirements, develop a plan of action, receive and manage resources,
implement their projects, and ensure quality and accountability. Such programmes could target
areas with potential for return. In areas with potential for return, community analysis could include
priority needs for when people start to return. Analysis on prioritisation could be shared with refugees
in camps.
the early 1990’s and since 2006. Many noted that when Al-Shabab captured Afmadow in 2009, large numbers of people ed to Kenya.
Some people were also said to have ed due to land disputes.
1.3 Livelihood and food security
Pastoralism is the dominant livelihood activity in the region. Other activities include petty trade, carpentry, farming, and charcoal burning
(Sedhuro, 2012; key informants).
Lower Juba is the largest producer of maize and sorghum in Southern Somalia, but ash oods in November and December 2011 have
damaged large portions of crops (FSNAU, 2012: 8). Additionally, due to a long dry spell between April and June 2012, the harvest was
signicantly below average (FSNAU, 2012a: 1). Between February and June 2012, 90,000 people in rural and urban areas were categorised
as in an emergency food security situation (FSNAU, 2012: 3). In interviews, key informants regularly stressed that food insecurity was a
major problem in Afmadow. Nutrition continues to be affected by the 2011 drought. Severe acute malnutrition rates for April-July 2012
were 25.1% in the Juba agro-pastoral region. In the Juba riverine and Juba pastoral regions the rates were 21.5% and 16% respectively
(Nutrition Cluster, 2012).
1.4 Public infrastructure and basic services
Key informants generally describe public services in Afmadow as of poor quality and insufcient to meet the needs of the population. The
lack of qualied staff in the health and education sectors is consistently highlighted. Several mentioned that qualied workers have been
moving to other areas of the country or to neighbouring countries to seek better opportunities and/or security.
There are seven health facilities in Lower Juba (Health cluster). In Afmadow, where there is a health centre and a hospital, key informants
stressed that access to health services is insufcient, facilities are in poor condition, and more should be done to address malnutrition.
There is one public primary school and one secondary school in Afmadow (local authorities), a number considered too low to meet the
needs of the school-aged population by key informants and the Somalia Education Cluster. This means that a number of children can
only access Quranic schools. In addition, key informants commonly raised the lack of learning material, as well as suitable WASH facilities.
Access to water in Afmadow, mostly through wells, is deemed good. On the other hand, sanitation is described as non-existent or extremely
poor, which is consistent with other assessments. FSNAU notes that poor WASH facilities increase the outbreak of diseases such as
measles and watery diarrhoea (FSNAU, 2012). A number of interlocutors stressed that roads were in a poor state.
2. Baidoa in Bay Region
2.1 Security, stability and humanitarian access
The capital city of Bay region (pop. 620,500) and one of Somalia’s largest cities, Baidoa was taken over from Al-Shabab in February 2012.
Yet, in July, OCHA reported that the group was still in control of most of the villages and rural areas surrounding the town (OCHA, July
2012: 11). The city that used to house Somalia’s interim parliament had been seized by Al-Shabab in January 2009 and was the second
most important base for the group, after Kismayo (Al Jazeera, 16 Aug. 2012; BBC, 22 Feb. 2012).
While the region is relatively secure, humanitarian access remains very restricted (OCHA, Aug. 2012). In November 2011, six out of seven
international organisations were banned from the region by Al-Shabab (OCHA, March 2012). Between February and June 2012, there
was limited insecurity and the situation was expected to remain the same (FSNAU, 2012). Key informants generally consider the security
to be average in Baidoa. The regional governor, deputy mayor and a few clan leaders mentioned the presence of unexploded ordnance
in the surroundings of the city and in rubbish dumps.
DURABLE SOLUTIONS | 43
Baidoa, located some 250 km southeast of the Ethiopian border, is home to 340,000 people, according to local authorities. Its population
is predominantly from the Rahaweyn, Digil and Mirie clan and sub-clans (SCWRW, 2012: 6).
2.2 Displacement
It is estimated that some 40,000 people are internally displaced in the region, while more than 80,000 people have ed to another country,
mostly Ethiopia, but also Kenya (OCHA, July 2012: 22; UNHCR statistics).
Local authorities assess that approximately 8,000 people are displaced in Baidoa district and hosted by the community or living in a
number of settlements (between 24 and 40, according to various authorities). Their estimates of internal displacement from Baidoa to
other districts in Somalia vary between 30,000 and 45,000. Figures of estimated refugees from Baidoa are much higher, ranging between
200,000 (mayor) to 250,000 (regional governor). Such estimates are most likely higher than real gures given that “only” 80,000 registered
refugees in camps originate from the region.
Key informants noted that 1991-1992, 2003, 2007-2008 and 2011-12 were key periods of displacement, driven by war, clan conict and
insecurity. For all periods except 2007-2008, drought was also identied as one of the causes of displacement.
2.3 Livelihood and food security
Baidoa is a major economic centre in southern Somalia, with strong economic links to other rural and urban centres in the region, notably
Mogadishu, Marka and Belet Xaawo. It is the trading centre for surrounding sorghum producing regions, Gedo, Bay, Bakool and Hiran. Main
economic activities in the town include business, casual labour, self-employment, and livestock and agricultural trade (FSNAU, 2009: 1). In
2009, 10% of Baidoa’s urban population was deemed very poor, 30% poor, 45% middle class and 15% better-off (FSNAU, 2009: 1). Key
informants ranked farming as the most important livelihood activity, followed by raising livestock, petty trades, trades and business activities.
Between February and June 2012, an estimated urban population of 45,000 and a rural population of 200,000 were categorised as in
food security crisis in the region (FSNAU, 2012: 3). Watery diarrhoea, cholera, and measles, in addition to poor WASH programmes and
substandard childcare facilities, make it challenging to control malnutrition rates (FSNAU, 2012:16).
2.4 Public infrastructure and basic services
Just as in Afmadow, key informants consider availability and quality of basic services unsatisfactory and stress the lack of qualied workers,
especially in the health and education sectors. Several mentioned that qualied teachers and health workers have left the district for other
parts of Somalia or other countries.
There is one hospital, nine functioning health posts and 11 mother and child health centres in Baidoa, according to the Somalia Health
Cluster. Key informants commonly said that the existing infrastructure is not sufcient to meet the needs of the population and neither
is the quality of the services provided.
Baidoa has one university and seven public primary and secondary schools. In addition, there are a number of private schools. Bay has
the second highest school enrolment in south-central Somalia, with boys outnumbering girls two-to-one (Education Cluster, 2011). A
majority of key informants deemed access and the quality of education in Baidoa average. Approximately a third felt it was insufcient,
mostly underlining that there should be more schools. Several also noted the lack of access to higher education.
Access to water in Baidoa city is considered good to average by key informants, even though a number underlined the need for a public
provision of clean water. Interlocutors considered sanitation infrastructure insufcient to non-existent. The regional governor highlighted
that rubbish collection needed to resume. The state of roads was described as inadequate and poor telecommunications and an erratic
provision of electricity as problematic.
44 | DURABLE SOLUTIONS
3. Belet Xaawo in Gedo Region
3.1 Security and humanitarian access
In late 2011, TFG troops started taking over Gedo region, bordering Kenya to the west and Ethiopia to the north, from Al-Shabab. In July
2012, OCHA reported that the TFG controlled most parts of the ve districts (Belet Xaawo, Ceel Waaq, Luuq, Doloow and Garbahareey)
and that humanitarian access had improved (OCHA, July 2012a). Access was still considered poor (OCHA, Aug. 2012). In November
2011, NGOs and UN agencies were banned from the region, hampering the provision of humanitarian assistance. Yet, some organisa-
tions, mostly local, have continued to deliver life-saving assistance (OCHA, Feb. 2012). In interviews, local authorities, clan and religious
leaders and representatives of civil society generally deemed the security good. A number of key informants underlined the presence of
unexploded ordnance in the district and that Al-Shabab had planted mines along the Kenyan border.
Gedo has an estimated population of 328,378, of which some 22,000 live in Belet Xaawo district (UNDP, 2005; D.C Belet Xaawo). Marehan
(Darod) is the dominant clan and Ali Dehere is the dominant subclan in Belet Xaawo (GSDO, 2012: 1).
3.2 Displacement
It is estimated that some 77,000 people are displaced in the region (OCHA, July 2012: 22). Local authorities and IDP leaders in Belet
Xaawo indicate that between 6,000 and 8,000 IDPs live in the locality, with the host community or in settlements. Some 124,000 refugees
in Kenya and Ethiopia come from Gedo. A majority are hosted in Dolo Ado and Dadaab camps.
Local authorities, elders and civil society organisations assess that up to 10,000 people from Belet Xaawo have been displaced to another
district and up to 10,000 people have become refugees. A number of key periods and drivers of displacement were listed, including the
civil war of the early 1990’s; ghting between Al-Ittihad, aiming at uniting Ethiopia’s eastern region with Somalia, and Ethiopian troops in
1996; inter-Marehan ghting from 2001 to 2004; the rise of the ICU in 2006; the drought from October 2010; and the invasion by Kenya
Defence Forces a year later.
3.3 Livelihood and food security
Gedo’s population mostly relies on livestock, farming, and petty trade. Many pastoralists lost their livestock because of erratic rains (OCHA,
Feb. 2012). In addition, farmers have had to use large portions of their harvest to repay debt from the 2011 drought, leaving them with
little to sustain themselves (OCHA, Dec. 2011: 1).
Gedo’s agro-pastoral community continues to struggle with food security and is largely dependent on aid because of the ongoing drought
and recurrent food shortages (OCHA, Dec. 2011 and Feb. 2012). In November and December 2011, ash oods damaged large portions
of crops. In addition, because of the long dry spell between April and June 2012, harvests were signicantly below average (FSNAU, 2012a)
.
Between April and July 2012, malnutrition rates in the North Gedo pastoral area were 28.4% and 22% in the North Gedo riverine area
(Nutrition Cluster, 2012). Malnutrition is linked to the interruption of humanitarian access and outbreaks of acute watery diarrhoea, cholera,
malaria, measles and whooping cough (OCHA, Dec. 2011: 2). Overall, food is of low dietary quality, childcare facilities are substandard,
and there is limited access to WASH programmes (FSNAU, 2012).
3.4 Public infrastructure and basic services
Similar to other locations, access to basic services and public infrastructure is considered inadequate and the lack of qualied staff is
commonly emphasised. Insecurity attributed to Al-Shabab is said to prevent access to basic services, especially as people might have to
walk long distances to reach public facilities.
Belet Xaawo has seven health facilities and one hospital. There are two mother and child health centres, four health posts and one mobile
clinic (GSDO, 2012: 3). Key informants generally deem the access to quality health services insufcient and note the lack of equipment,
medicine and ambulances.
In 1991, following the outbreak of civil war, schools became the responsibility of community education committees. Even though these
committees now offer teacher-training opportunities in Belet Xaawo, there are not enough qualied teachers to meet the needs of school-
aged children. Many children, particularly in rural areas, only have access to Quranic schools. There are no secondary schools in the
district, an issue commonly highlighted by key informants (GSDO, 2012: 3).
As of December 2011, less than 20% of the population had access to a water source (OCHA, Dec. 2011: 1). Key informants rated access
to water, usually through communal boreholes, as average. Yet, they noted limited access to clean water.
Sanitation was deemed insufcient. Roads were described as impassable. Key informants regularly stressed the lack of electricity, post
ofces, banking services and Internet access.
DURABLE SOLUTIONS | 45
b) Impact of displacement and of an eventual large-scale return of refugees and IDPs
1. Spontaneous returns of refugees and IDPs
1.1 Impact of displacement:
In addition to mentioning that people who ed generally abandoned their land and other assets and left their houses unoccupied, key
informants stressed that such departures had meant a loss of qualied workers. Some ed, while others who stayed lost their jobs and
moved in search of employment, as schools and health centres were left unused. Economic activity diminished, due to a decrease in the
population, looting and violence. Still, a number of key informants pointed out that members of the community had continued to benet
from remittances from refugees abroad.
1.2 Spontaneous returns:
Key informants in the three districts studied said that small numbers of refugees and IDPs had been returning spontaneously. In Afmadow,
local authorities and clan leaders estimated that between 3,000 and 11,000 IDPs and up to 15,000 refugees had returned. In Baidoa,
estimates of returning IDPs varied between 500 and 1,500, while numbers of refugees ranged between 100 and 500. Most returned to
agro-pastoral areas in the vicinity of Baidoa. Yet, according to IDP leaders, a few hundred returned to the town of Baidoa and settled in
IDP camps. In Belet Xaawo, gures ranged between 500 to 2,000 for IDPs and 1,000 to 3,000 for refugees.
In all districts, key informants mentioned that IDPs and refugees had been in touch with the community, some also coming back to look
after their land and other assets. In Belet Xaawo, it was also mentioned that some refugees have visited to share their skills and knowledge
with people in the locality.
1.3 Reintegration of returnees:
Former IDPs and refugees have faced specic challenges related to the return and reintegration in the three districts covered, notably
with respect to the restoration of their land and other properties. Houses and land left behind, that have often been occupied, are said
to be difcult to regain. This has created tensions and conicts between returnees and those who stayed behind. It was often stressed
that those who still had access to their land had relatives looking after it. According to key informants, other assets have commonly been
looted or, in the case of livestock, have been depleted during the recent drought. Shelters that have not been occupied are usually in bad
condition and xing them has been challenging, since returnees have limited means.
In addition, key informants in all districts note that spontaneous returnees have faced a lack of access to basic services, food and job
opportunities. In Belet Xaawo, a number of key informants mentioned that returnees were sometimes discriminated against when trying
to access justice. Tribalism was, in many instances, highlighted as an obstacle to return and reintegration
.
1.4 Return mechanism:
While spontaneous returns have been occurring in the three districts, no ofcial mechanism to facilitate the reintegration of returnees has
been established. In Belet Xaawo, key informants reported that elders have been helping returnees re-acquire their land and solve conicts
through arbitration. In all locations, it was stressed that such a mechanism should be put in place by the government, in collaboration with
UN Agencies and humanitarian organisations, notably looking into access to land and housing.
2. Impact and pre-conditions for a sustainable large-scale return
2.1 Projected impact and challenges
In all districts, the return of refugees and IDPs in large numbers would be expected to create tensions and competition over scarce
resources, access to overstretched basic services and limited job opportunities. Given that food insecurity is a common occurrence in the
districts covered, returnees’ access to food would likely be an additional challenge. In all districts, it was noted that, for the time being,
security and safety would remain a challenge.
In Afmadow, the restoration of land and other properties, including livestock, was seen as especially problematic and many warned that
this could lead to clan conicts. Most key informants believed that returnees would no longer have access to their land and houses,
unless a relative had been looking after them. All believed that livestock left behind would have been taken over and would be impossible
to recognise. However, key informants also noted that existing conict resolution mechanisms could be put to use, most commonly citing
community and religious leaders and clan elders. In some cases, Islamic leaders and local authorities were also mentioned.
46 | DURABLE SOLUTIONS
Similarly, in Belet Xaawo, in the absence of means and proof of ownership, restoration of land and housing is expected to be problematic,
leading to food insecurity and a high number of conicts. It was noted that existing conict mechanisms – where clan elders, local and
religious leaders settle disputes through swearing of witnesses and neighbours and convening communal meetings – could contribute
to easing tensions.
The perspective was more optimistic in Baidoa town where it was largely felt that returnees would have access to their land and that
conicts would be isolated cases that could be resolved legally. This could be partially explained by the fact that a number of returnees to
urban areas are likely to earn their living through an activity requiring limited access to land, such as business or casual labour. Restoration
of shelter, livestock and other belongings was deemed unlikely.
While key informants in Baidoa did not expect problems related to discrimination or access to justice, the perception in Belet Xaawo
was different. Several people warned against discrimination towards returnees, who would be perceived as “foreigners, cowards, and
betrayers”. Such discrimination could lead to biases in favour of those who stayed and towards fellow clansmen in the exercise of justice.
2.2 Pre-conditions for sustainable and peaceful returns
The regional governor of Bay stressed that before refugees or IDPs can return, reasons that caused their displacement must no longer
prevail and security must be of a lasting character. So as to make large-scale returns possible and sustainable, key informants in all districts
pointed out that access to quality basic services, land, shelter and water should be improved and food security and livelihood opportunities
bolstered. It was also suggested that refugees should also be equipped with appropriate skills while in exile.
2.3 Support by humanitarian organisations
Several key informants stressed that humanitarian organisations could help communities prepare for a sustainable return of IDPs and
refugees by contributing to improvement of the quality and access to basic services, building infrastructure (notably roads and electricity),
enhancing food security, supporting livelihood activities and skills training and the local economy, as well as activities aimed at restoring
and consolidating peace and security. In Baidoa, it was suggested that aid organisations could contribute to identifying land for returnees,
if necessary, and should provide direct support to people returning, such as food and shelter.
2.4 Useful skills and tools
In all districts, key informants highlighted a critical shortage of qualied staff, notably due to unattractive wages encouraging people to look
for job opportunities in Kenya or Ethiopia, but also because of the insecurity that forced many to ee and, in the absence of a functioning
administration, a lack of training structures. The decit of teachers, doctors and other health professionals was considered especially grave.
The need for administrators was commonly highlighted, while engineers and business people capable of contributing to the development
of the local economy were also mentioned. In general, people with skills were described as lacking and deeply necessary. Similarly, it is
generally agreed that people coming back with assets such as farming tools, livestock and sewing machines would have an easier time
earning their living again and would contribute to the local economy.
3. Observations
Clear benchmarks to ensure that sustainable security is in place in areas of return, that returnees would have access to land, livelihood
opportunities and essential services, and that humanitarian organisations have secure access, seem crucial before a sustainable
repatriation program can take place. None of the three localities studied, all recently retaken from Al-Shabab, situated in regions hosting
signicant IDP populations and home to signicant numbers of refugees, would currently meet basic standards.
In the three locations, similar to most of south-central Somalia, after two decades of conict and instability and in the absence of a
functioning government, public infrastructure and basic social services are very limited and, when available, often of unsatisfactory quality.
The population has been surviving in difcult conditions and has commonly been using the land and assets of people who ed.
Small numbers of refugees and IDPs are reported to have returned spontaneously to the districts studied and have generally faced
challenges with the restoration of their land and other properties, which has, in some instances, created tensions and conicts between
returnees and those who stayed behind. According to key informants, assets have often been looted or, in the case of livestock, have
been depleted during the recent drought. Shelters that have not been occupied are usually in poor condition and returnees have limited
means to x them. Returnees also lack livelihood opportunities, food and access to basic services.
Authorities in Kenya have discussed the creation of “Jubbaland” in southern Somalia as a safe zone intended to curb the ow of refugees
into Kenya as well as allow for the repatriation of refugees to southern Somalia. Yet, as highlighted by the regional governor of Bay, before
IDPs and refugees return, reasons that caused their displacement should no longer prevail and security must be of a lasting nature.
Premature large-scale returns would probably be unsustainable and lead to increased tensions and competition over scarce resources.
DURABLE SOLUTIONS | 47
Key informants pointed out that to make large-scale returns possible and sustainable, food security and access to quality basic services,
shelters, water and livelihood opportunities must be improved. Restoration of and access to land will be a crucial and sensitive matter,29
especially as a large part of the potentially repatriating refugee population will substantially rely on farming for their livelihood, and hence
require access to sizeable land.
Struggle for land has been a central element to the Somali conict. Land grabbing and reallocation has been common in the last decades
(Displacement Solutions, 2008: 9). Even though returnees’ need for land will vary according to their livelihood activities,30 access to land
will be essential in all cases, at least for housing purposes. Clear, fair and efcient restitution mechanisms, allowing for the restoration of
land and other property or for appropriate compensation, will also be necessary so as to avoid creating new conicts. In addition, certain
customary rules discriminate against women and minority clans, which means that the situation of women-headed households and
minorities would have to be considered carefully (Displacement Solutions, 2008: 13).
Informants also suggested that refugees should be equipped with practical skills while in exile. In all districts, key informants highlighted
a critical shortage of qualied staff, notably due to large-scale departures - a number ed, while some of those who remained, lost their
jobs, as schools and health centres were left unused, and ended up seeking job opportunities elsewhere. Unattractive wages and a lack
of training structures have also encouraged a number of qualied people to move to Kenya or Ethiopia.
The dearth of teachers, doctors and other health professionals is reportedly especially grave. The need for administrators was also
commonly highlighted, while the need for engineers and business people capable of contributing to the development of the local economy
also emerged. In general, people with skills were described as acutely needed. Returnees with assets, such as farming tools, livestock
and sewing machines, would have an easier time earning their living again, which, in addition to contributing to the local economy, would
facilitate their reintegration.
Several key informants stressed that humanitarian organisations could help communities prepare for a sustainable repatriation by
contributing to improving quality and access to basic services, building infrastructure (notably roads and electricity), enhancing food
security, supporting livelihoods and skills training and the local economy, as well as activities aimed at restoring and consolidating peace
and security. In addition to providing direct support to people returning in the form of food and shelters, aid organisations could help the
setting up mechanisms aimed at preventing and addressing land-related tensions.
Given the socio-economic situation of the population in Somalia, activities and aid targeting returnees could contribute to creating tensions
and discrimination against returnees, unless appropriate measure are also taken to improve the living standards of the overall population.
29 For more on land, property and housing in Somalia, see NRC, UNHCR and UNHABITAT (2008). “Land, Property and Housing in Somalia.”; Displacement Solutions (2008). “Housing, Land and
Property Rights in the South Central Somalia; Preliminary Assessment and Proposed Strategies for Expanded Approaches.”; Unruh JD (1995). “Pastoralist resource use and access in Somalia: a
changing context of development, environmental stress, and conflict.” In: Sorenson J (ed.) Disaster and Development on the Horn of Africa. Macmillan, London.
30 Agriculturists need access to farm land, agro-pastoralists would need access to a combination of farm and grazing land, while pastoralists are likely to need seasonal access in different
areas. Business-people, casual labourers or specialised workers might only need access to land for their housing and possibly their shops.
48 | DURABLE SOLUTIONS
DURABLE SOLUTIONS | 49
8 Concluding remarks
1. This study was undertaken as an attempt to better understand how displaced Somalis view their situation, envisage their future and, if
their repatriation ever became possible, what would make it sustainable. Given its relatively small sample, the research was not designed
to allow statistical projections, but to permit pertinent observations. It is not a planning document, but amounts to an analytical input
into further planning on durable solutions and repatriation. For planning purposes, a better understanding of the overall situation and of
land tenure, property access and ethnic geographies in areas of potential return in south-central Somalia will be essential. Similarly, such
planning will require precise household level data on refugees living in camps. The verication exercise currently taking place in Dadaab
represents a valuable contribution in that respect.
2.
A fairly signicant proportion of Somali refugees in camps would consider returning to Somalia when conditions become conducive
for a sustainable repatriation, especially those who have ed recently. At this point in time, a majority of refugees in camps assess that the
reasons that led to their displacement still prevail, and thus would not consider returning in the near future. They should not be pushed
do to so by inadequate or deteriorating security and living conditions in the refugee camps.
3. Repatriation is not necessarily the optimal solution for all refugees and after decades in exile, a signicant proportion is likely to have
limited reintegration prospects, even if conditions for return become conducive. Lack of access to land, which is closely linked to livelihood
opportunities, and limited personal connection to the country appear as key impediments to reintegration. The longer refugees have been
in exile, the less likely they are to still have access to land. Similar observation can be made for minority groups and youth. The latter are
also less likely to feel personally connected to Somalia, a country that, in many cases, they have never visited. For different reasons, the
return and reintegration of people who have had traumatic experiences also appears unlikely. This brings to light the fact that no response
will be adequate for all refugees, thus the need to be creative in the search for durable solutions and further explore pathways to
intermediate integration, local integration and mobility opportunities that allow for residency and/or work permits. Further research should
be undertaken to inform strategies for specic groups and how they impact on return.
4. In order to make returns sustainable in a country destroyed by decades of conict, careful planning will be necessary, in close
collaboration with local authorities and clan leaders. Experience has shown that durable security, basic services and humanitarian access
are minimal prerequisites for durable repatriation, which should not be seen as a simple logistical movement of people, but as a part of
wider development and peacebuilding projects. Premature or hasty mass returns could hinder peacebuilding, reconstruction and sustain-
able reintegration. Without clear benchmarks to ensure that sustainable security is in place in areas of return, that essential services and
durable livelihood opportunities are available, returnees have livelihood opportunities, and that humanitarian organisations have secure
access, a repatriation programme would likely be unsustainable. Given the socio-economic situation of the population in Somalia, activities
and aid targeting returnees could cause tension, unless appropriate measure are also taken to improve the living standard of the overall
population. Experience also suggests that smaller and incremental population movements, coupled with assistance and monitoring of
returnees can help make return and reintegration sustainable.
5.
Restoration of and access to land will be a crucial and extremely sensitive matter, especially as a large part of the potentially repatriating
refugee population will rely on farming and agro-pastoralism for their livelihood, hence requiring signicant access to land. In the absence
of such access, refugees livelihood opportunities are likely to be narrow, thus their reintegration prospect limited. A number of returnees
who will earn their living through activities that are not land-based, such as business or manual labour, will have more limited needs for
land. Special attention should be given to ensuring that fair and efcient restitution system is in place to address matters of restoration and
compensation, so as to avoid creating further conicts. In addition, the situation of women and minorities deserves special attention, as
certain customary rules discriminate against them, which could further limit their reintegration prospects (Displacement Solutions, 2008: 13)
.
6.
Helping people in camps develop skills should enhance their prospects for durable return, by enabling them to reintegrate and contribute
more substantially to the reconstruction of Somalia. This should also enhance their livelihood opportunities in the refugee camps, as well
as their capacity to integrate locally or in a third country. Considering longer-term skills transfer and mentoring by aid organisations could
be benecial. Focusing on skills that are deemed necessary in Somalia is advisable, especially as these qualications would also be useful
in exile. In Somalia as in the camps, the decit of qualied teachers appears especially critical. Health professionals, administrators and
other trained professionals would also be extremely valuable. In addition, specialised knowledge in elds related to construction are likely
to be useful.
50 | DURABLE SOLUTIONS
Bibliography
Abdullahi, Ahmednasir M. (1993). “Protection of Refugees Under International Law and Kenya’s Treatment of Somali Refugees: Compliance or Contrary?”
——— (2008). Gérer les indésirables : des camps de réfugiés au gouvernement humanitaire. Paris: Flammarion.
Al Jazeera (16 Aug. 2012). “Timeline: Somalia’s historic highlights.”
Ambroso, Guido (1993). “Pastoral society and transnational refugees: population movements in Somaliland and eastern Ethiopia 1998-2000”
UNHCR Brussels. Working Paper No. 65.
ARRA (2010). “ARRA Update.” Vl IV (17), Oct-Dec. 2010, Administration for Refugees and Returnee Affairs. Available at: http://www.arra.
org.et/images/pdfs/cover/update23.pdf
——— (2009). “The state of refugees in Ethiopia”. http://www.arra.org.et/index.php/faq/53-the-state-of-refugees-in-ethiopia
Banki, Susan (2004). “Refugee Integration in the Intermediate Term: A Study of Nepal, Pakistan, and Kenya.” Geneva: Evaluation and
Policy Unit, United Nations High Commissioner for Refugees.
BBC (31 May 2012). “Somalia forces capture key Al-Shabab town of Afmadow.”
——— (22 Feb. 2012). “Somalia Al-Shabab militant base of Baidoa captured.”
——— (17 Oct. 2011). “Kenya’s Incursion into Somalia Raises the Stakes.”
——— (19 Nov. 2011). “Ethiopian troops cross border into Somalia.”
——— (31 Dec. 2011). “Ethiopian troops capture Beledweyne from Somalia militants.”
Bradbury, Mark, and Sally Healy (2010). “Endless War: A Brief History of the Somali Conict.” Accord, no. 21
Campbell, Elizabeth, Jeff Crisp and Esther Kiragu (2011). “Navigating Nairobi; A review of the implementation of UNHCR’s urban refugee
policy in Kenya’s capital city.” Geneva: United Nations High Commissioner for Refugees Evaluation and Policy Analysis Unit.
Chimni, B.S. (1993). “The Meaning of Words and the Role of UNHCR in Voluntary Repatriation.” International Journal of Refugee Law 5, no. 3: 442-60.
Collin, John S. (1996). “An Analysis of the Voluntariness of Refugee Repatriation in Africa.” University of Manitoba.
Colson, Elizabeth (1999). “Gendering Those Uprooted by ‘Development’.” In Engendering Forced Migration : Theory and Practice, edited
by Doreen Marie Indra, 23-39. New York: Berghahn Books.
Constitution of Kenya, 2010.
Convention Governing the Specic Aspects of Refugee Problems in Africa, Organization of African Unity, 1969.
Convention relating to the Status of Refugees, 1951.
Convention on the Rights of the Child, 1989.
Cooper, Elizabeth (2005). “What Do We Know About out-of-School Youths? How Participatory Action Research Can Work for Young
Refugees in Camps.” Journal of Comparative Education 35, no. 4: 463-77.
Crisp, Jeff (2003). “No Solutions in Sight: The Problem of Protracted Refugee Situations in Africa.” In New Issues in Refugee Research,
edited by Evaluation and Policy Analysis Unit, 38: UNHCR.
Crisp, Jeff, and Karen Jacobsen (1998). “Refugee Camps Reconsidered.” Forced Migration Review, no. 3: 27-31.
Dippo, Don, Aida Orgocka, and Wenona Giles. “Reaching Higher: The Provision of Higher Education for Long-Term Refugees in Dadaab
Camps, Kenya.” Centre for Refugee Studies, York University, 2012.
Displacement Solutions (2008). “Housing, Land and Property Rights in the South Central Somalia; Preliminary Assessment and Proposed
Strategies for Expanded Approaches.”
DRC, UNHCR, and FEG consulting (2012). Building the Self-Reliance of Urban Refugees; ‘Living on the Edge’ in Nairobi. Nairobi: Danish
Refugee Council, United Nations High Commissioner for Refugees and FEG consulting.
DRC (2011). “Preliminary impact assessment of DRC’s livelihood projects.” Danish Refugee Council, Ethiopia Programme.
Don Bosco (2011). “Don Bosco VTC Kakuma - Prole and Statistics .”
Dryden-Peterson (2011). “Refugee Education; a Global Review.” Toronto: Ontario Institute for Studies in Education, University of Toronto.
Education Cluster (2011). “Overall Enrolment of Learners.”
Enghoff, Martin et al. (2010). “In Search of Protection and Livelihood; Socio-Economic and Environmental Impacts of Dadaab Refugee
Camps on Host Communities.” Nairobi: Royal Danish Embassy, Norwegian Embassy and Republic of Kenya.
Essed, Philomena, Georg Frerks, and Joke Schrijvers (2004). “Introduction: Refugees, Agency and Social Transformation.” In Refugees
and the Transformation of Societies. Agency, Policies, Ethics and Politics, edited by P. Essed, Frerks, G. and Schrijvers, J., 1-16.
New York/Oxford: Berghahm Books.
Farah, Nuruddin (1998). “A Country in Exile.” World Literature Today 72, no. 4: 713-15.
FSNAU (2012). “Special Brief – Post Deyr 2011/12 Analysis.” Food Security and Nutrition Analysis Unit – Somalia.
——— (2012a). “Somalia Post-Gu 2012 Food Security and Nutrition Outlook – August to December 2012.”
——— (2009). “Livelihood Baseline Prole; Baidoa Urban.”
GoE (2004). Refugee Proclamation. No. 409/2004. Government of Ethiopia.
——— (2003). Ethiopian Nationality Proclamation. No. 378/2003. Government of Ethiopia
GoK (2011). Kenya Citizenship and Immigration Act. Government of Kenya.
——— (2006). The Refugee Act of Kenya. Government of Kenya.
DURABLE SOLUTIONS | 51
GSDO (2012). “Final Report: Belet Xaawo district.” Gedo Social Development Organisation.
Gundel, Joakim (2002). “The Migration-Development Nexus: Somalia Case Study.” International Migration 40, no. 5: 255-81.
Harrell-Bond, Barbara E (1986). Imposing Aid : Emergency Assistance to Refugees. Oxford; New York: Oxford University Press.
Horst, Cindy (2008). “A Monopoly on Assistance: International Aid to Refugee Camps and the Neglected Role of the Somali Diaspora.”
Afrika Spectrum 43, no. 1: 121-31.
——— (2006). Transnational Nomads. How Somalis Cope with Refugee Life in the Dadaab Camps of Kenya. Oxford et New York: Berghahn Books.
——— (2006a). “Buus Amongst Somalis in Dadaab: The Transnational and Historical Logics Behind Resettlement Dreams1.” Journal
of Refugee Studies 19, no. 2 : 143-57.
HRW (July 2012). “Human Rights Watch’s response to the British Foreign and Commonwealth Ofce’s Quarterly Update: Somalia.”
——— (March 2012). “Kenya: Somalia Unsafe for Refugees to Return.” Nairobi: Human Rights Watch.
Hyndman, Jennifer (2000). Managing Displacement : Refugees and the Politics of Humanitarianism, Borderlines. Minneapolis, MN ;
London: University of Minnesota Press.
Hyndman, Jennifer, and Bo Viktor Nylund (1998). “UNHCR and the Status of Prima Facie Refugees in Kenya.” International Journal of
Refugee Law 10, no. 1-2: 21-48.
IRIN (24 Nov. 2009). “Somalia: Mass exodus as militia takes control of southern town.”
——— (26 July 2007). “Ethiopia-Somalia: New Camp Opened for Somali Refugees.”
Jacobsen, Karen (2005). The Economic Life of Refugees. Bloomeld, CT: Kumarian Press.
Jansen, Bram J. (2011) “The Accidental City; Violence, Economy and Humanitarianism in Kakuma Refugee Camp, Kenya.” PhD Thesis,
Wageningen University.
Jureidini, Ray. “Mixed Migration through Somalia and across the Gulf of Aden.” Nairobi, Cairo: Mixed Migration Task Force Somalia and
Yemen / American University in Cairo, 2010.
Kagwanja, Peter and Monica Juma (2008). “Somali refugees: Protracted exile and shifting security frontiers”, Protracted Refugee
Situations, ed. Gil Loescher, James Milner et al. New York: United Nations University Press: 214-247.
Kibreab, Gaim (1999). “Revisiting the Debate on People, Place, Identity and Displacement.” Journal of Refugee Studies 12, no. 4: 384-410.
Kirkby, John, Ted Kleist, Georg Frerks, Wiert Flikkema, and Phil O’ Keefe (1997). “UNHCR’s Cross Border Operation in Somalia: The Value
of Quick Impact Projects for Refugee Resettlement” in Journal of Refugee Studies Vol. 10, No. 2 1997.
Lindley, Anna (2011a). “Between a Protracted and a Crisis Situation: Policy Responses to Somali Refugees in Kenya.” Refugee
Survey Quarterly 30, no. 4: 14-49.
——— (2007). “Protracted displacement and remittances: the view from Eastleigh, Nairobi.” in New Issues in Refugee Research
Reasearch Paper #143.
Lindley, Anna and Haslie, Anita (2011). “Unlocking Protracted Displacement; Somali Case Study.” In Working Paper Series. Oxford: Refugee
Studies Centre.
Long, Katy (2011). “Permanent Crises? Unlocking the Protracted Displacement of Refugees and Internally Displaced Persons;
Policy Overview.” Oxford: Refugee Studies Centre.
——— (2010). “Home Alone? A Review of the Relationship between Repatriation, Mobility and Durable Solutions for Refugees.” In Policy
development and evaluation service: UNHCR.
——— (2010a). “No entry! A review of UNHCR‟s response to border closures in situations of mass refugee inux.” In Policy development
and evaluation service: UNHCR.
Malkki, Liisa (1995). “Refugees and Exile: From “Refugee Studies” to the National Order of Things.” Annual Review Anthropolgy 24: 495-523.
Markos, Kibret (1997). “The Treatment of Somali Refugees in Ethiopia under Ethiopian and International Law.” International Journal of
Refugee Law 9, no. 3: 365-391.
Mbogo, Steve (4 Aug. 2012). “Agencies reject plan to relocate Dadaab Refugees” In The East African newspaper.
Menkhaus, Ken (2010). “Stabilisation and Humanitarian Access in a Collapsed State: The Somali Case.” Disasters 34, no. S3.
——— (2008). “Current Political and Socio-Economic Dynamics in Somalia.” UNICEF Somalia.
Montclos (Pérouse de), Marc-Antoine (1999). “Exodus and reconstruction of identities: Somali “minority refugees” in Mombasa.”
(Unpublished? Published in French?).
NRC (2012). “New Report on Education in Dadaab.” Norwegian Refugee Council.
Nutrition Cluster (2012). “May-July Survey Findings.”
OCHA (Aug. 2012). “Somalia Humanitarian Access Map.”
——— (July 2012). “Mid-Year Review of the Consolidated Appeal for Somalia 2012.”
——— (July 2012a). “Gedo Fact Sheet.”
——— (March 2012). “Bay Fact Sheet.”
——— (Feb. 2012). “Gedo Fact Sheet.”
——— (Dec. 2011). “Gedo – Situation Analysis.”
——— (2002). “A study on minorities in Somalia.”
Omata, Naohiko (2012). “Repatriation and Integration of Liberian Refugees from Ghana: The Importance of Personal Networks in the
Country of Origin.” Journal of Refugee Studies, no. 23.
52 | DURABLE SOLUTIONS
PANA (30 March 2012). “Kenya urged not to return refugees to Somalia.”
PRM (2011). “Kenya Assessment; Evaluation of Humanitarian Assistance to Colombian Refugees in Ecuador and Somali Refugees in
Kenya: Supporting Protracted Refugees’ Livelihoods, Protection Space and Self-Reliance.” Washington: United States Department
of State, Bureau of Population, Refugees, and Migration.
RCK (2012). “Asylum Under Threat: Assessing the protection of Somali refugees in Dadaab refugee camps and along the migration
corridor.” Nairobi: Refugee Constorium of Kenya.
RMMS. “Country prole: Ethiopia.” Regional Mixed Migration Secretariat. Accessed on 22 Sept. 2012, http://www.regionalmms.org/ index.php?id=12
Ryle, J. (1992). “Notes on the Repatriation of Somali Refugees from Ethiopia.” Disasters, 16: 160–168.
SCWRW (2012). “Final Report: Baidoa district.” Somali Children Welfare and Rights Watch.
Sedhuro (2012). “Final Report: Afmadow district.” Sean Deveruxe Human Rights Organisation.
The Star (15 Sept. 2012). “New camp for refugees.” http://www.the-star.co.ke/local/rift-valley/94565-new-camp-for-refugees
UNDESA. “African Statistical Yearbook 2011.” United Nations Department of Economic and Social Affairs, 2011.
UNDP. “Status of MDGs in Somalia.” http://www.so.undp.org/index.php/Millennium-Development-Goals.html.
UNHCR (2012). “Chap. 3: In Search of Solidarity.” In The State of the World’s Refugees.
——— (2012a). “World’s Biggest Refugee Camp 20 Years Old.” 21 Feb. 2012.
——— (2012b). “Melkadida – Camp Snapshot.”
——— (2012c). “Kobe– Camp Snapshot.”
——— (2012d). “Kambioos Camp Prole – April 2012.”
——— (2012e). “Bokolmanyo– Camp Snapshot.”
——— (2012f). “Sub-Ofce Kakuma Operations.”
——— (2012g). “Hiloweyn – Camp Snapshot.”
——— (2012h). “Kakuma camp in Kenya surpasses its 100,000 capacity.”
——— (2012i). “Bur-Amino– Camp Snapshot.”
——— (2012j). “ Dollo Ado Fact Sheet June 2012.”
——— (2012k). “Somali Refugee Prepares for the Long Haul in Ethiopia.”
——— (Sept. 2012). “Dadaab Statistical Review.”
——— (13 Sept. 2012). “Kakuma: New Arrivals Registration Trends by Date – 2012”
——— (April 2012). “Situation Report: Somali Border Regions / Gedo, Lower and Middle Juba.”
——— (April 2012a). “Frequently Asked Questions About Resettlement.”
——— (2011). “Global Report.”
——— (2011a). “UNHCR Education Fact Sheets – Fall 2011.”
——— (2010). “UNHCR Global Appeal 2010-2011.”
——— (2009). “UNHCR starts moving Somali refugees to a new camp in Ethiopia.”
——— (2004). “Protracted Refugee Situations.” edited by Executive Committee of the High Commissioner’s Programme.
——— (2002). “UNHCR Statistical Yearbook.”
——— (1995). “Report of the High Commissioner for Refugees.” Fiftieth Session Supplement No.12 (a/50/12).
——— (Dec. 1994). “Review of UNHCR’s Kenya-Somalia Cross-border Operation.” EVAL/CROS/14, UNHCR Evaluation Reports.
Available at: http://www.unhcr.org/3bd4228b4.html
——— “Statistical Online Population Database.”
UNHCR Statistics. “Refugees in the Horn of Africa: Somali Displacement Crisis; Information Sharing Portal.” http://data.unhcr.org/horn-of-africa/regional.php.
UNHCR, and Education Partners (2011). “Joint Strategy for Education in Dadaab 2012-2015.”
UNHCR, UNICEF, and Education Partners (2012). “Joint Assessment of the Education Sector in Kakuma Refugee Camp.”
UNICEF/UNHCR (2012). “UNICEF/UNHCR Education Update July 2012.”
UN News Centre (2010). “UN Welcomes Ethiopian Policy to Allow Eritrean Refugees to Live Outside Camps.” 10 Aug. 2010.
Verdirame, Guglielmo, and Barbara E. Harrell-Bond (2005). Rights in Exile : Janus-Faced Humanitarianism, Studies in Forced Migration.
New York: Berghahn Books.
Waldron, Sidney, and Naima A. Hasci (1995). “Somali Refugees in the Horn of Africa.” In Studies on emergencies and disaster relief:
Refugee Studies Programme, University of Oxford.
WFP, UNHCR, ARRA (2010). “Ethiopia Joint Assessment Mission; Major Findings & Recommendations.” World Food Programme, United
Nations High Commissioner for Refugees and Administration for Refugee and Returnee Affairs.
Zimmerman, Susan (2009). “Irregular Secondary Movements to Europe: Seeking Asylum Beyond Refuge, and the Links between These
Stages.” Journal of Refugee Studies 22, no. 1: 74-96.
DURABLE SOLUTIONS | 53
Annex A: Methodology
(Excerpt from the research proposal)
The methodology outlined below focuses on the collection of primary data, even though the nal research will include references to relevant
literature. In addition, as a number of organisations and institutions were consulted during the preparation of this proposal (c.f., footnote
6), they will not be interviewed again during the gathering of data, unless necessary. However, these interviews will be integrated in the
nal report, when appropriate.
The case study on the repatriation from Jijiga to Somaliland will constitute a separate project. It will be conducted through a literature
review and interviews with key informants in Jijiga and Somaliland.
1. Data collection in the eld
The methodology proposed for the eldwork combines qualitative and quantitative tools and approaches.
a) Qualitative data collection
Qualitative assessments will comprise of focus group discussions in refugee camps and interviews with key informants in selected communi-
ties of return. Supplementary reports and information will also be collected from eld-based organisations and reviewed during the eldwork.
i. Focus group discussions (FGD): Some 65 FGDs will be conducted in the four main camp locations (Dadaab, Kakuma, Dolo Ado and
Jijiga) with refugees drawn from diverse populations in the camps, i.e., youth, women, minority groups, people considered vulnerable
(single-headed household, handicapped, chronic disease, poorer, etc.), entrepreneurs, incentive workers, unemployed adults, recent
arrivals (2011 and onwards), arrivals from 2006 and onwards and refugees who have been there for over a decade, camp leaders,
incl. women representatives and elders (c.f., “Interviews” sheet of Planning.xls for planning per camp). Each group will consist of a
maximum of 6 refugees.
Steps for the selection of FGD participants:
- In each location, the team leader will inform refugee leaders of the planned FGD one week in advance
- Refugee leaders will be asked to mobilise participants for the scheduled meetings
- Participants should always include non DRC beneciaries. Where DRC conducts activities, a small number of DRC
beneciaries should also be included.
ii. Interviews with key informants: In three selected districts of return, partners from the child-monitoring network will conduct interviews
with key informants. The latter include community leaders, elders, clan leaders, local authorities, service providers (health and
education), entrepreneurs and civil society organisations.
The suggested communities of return have been selected based on the areas of origin of refugees in camps, as it is possible to assume
that a number of refugees would be repatriating to their area of origin, and based on humanitarian access. The suggested districts are
Afmadow in Lower Juba, Baidoa in Bay, and Baardheere in Gedo.
Selection of key informants: Child-monitoring partners will be asked to identify key informants in the selected communities, based on
the enumeration above.
b) Quantitative data collection
Quantitative data will be collected through semi-structured interviews with 360 heads of refugee households (HH). It is proposed that inter-
views be conducted with heads of household rather than individuals, as they are likely to elicit more information about other members of
the household. In addition, it is likely that the decision to repatriate or not, or to leave a number of family members in a camp while others
return, will be a family decision and not an individual one, as families are likely to “pool their resources for mutual benet” (PRM, 2011: 17).
The household interviews will be conducted with the head of family present in the household at the time of the interview. If the two HH
heads are present, they will be interviewed together. Some 200 interviews will be conducted in Kenya and 160 in Ethiopia (c.f., “Interviews”
sheet of Planning.xls):
- In Dadaab’s ve camps, home to approx. 133,000 Somali households, 160 interviews will be conducted:
i. In Hagadera, Dagahaley and Ifo camps, 40 HH per camp, living in four different blocks will be interviewed (10 HH per block, at least
one block will be located at the periphery of the camp and one at the centre).
54 | DURABLE SOLUTIONS
ii. In Ifo 2 (East and West), Kambioos and Buramino, 20 HH per camp, living in 2 different blocs will be interviewed (10 HH per block,
one at the periphery of the camp, the other at the centre).*
- In Kakuma camp, home to approx. 9,500 Somali HH, 40 interviews in 4 different blocks and 6 focus group discussions are foreseen.
- In four of Dolo Ado’s ve camps (Bolkolmanyo, Melkadida, Hilaweyn and Buramino), home to approx. 30,550 Somali households,
80 interviews will be conducted. In each camp, 20 head of HH living in 2 different blocks of the camp will be interviewed.**31
- In Jijiga’s three camps, home to some 7,100 HH, 80 interviews will be conducted: 40 in Kebribeyah, 20 in Aw-Barre, and 20 in
Sheder. It is planned to conduct a relatively high number of interviews in Kebribeyha (given the size of the population) as the population
is expected to be somewhat more varied than in other Ethiopian camps.
By collecting data in all Dadaab, Kakuma and Jijiga camps, and most of Dolo Ado camps, the sample should include refugees from all
periods of arrival, from rural and urban areas, with differing levels of wealth, age, gender and clan afliation. For example, while Ifo’s old
residents mostly came from rural areas, many of Hagadera’s old residents came from urban areas (PRM, 2011: 19). It is also suggested
to gather information in all four main camp locations in Kenya and Ethiopia to verify how varying political and socio-economic conditions
in different locations affect the perspective of refugees with respect to durable solutions.
32
Steps for the selection of HH:
i. Prior to the start of the data collection, two or four blocks will be randomly selected in each camp, using the available maps. At least
one will be located at the periphery of the camp and one, near the centre.
ii. In the selected blocks, interviewers will conduct the rst interview at a randomly selected starting point. Once an interview is completed,
interviewers will leave the house, turn right, and stop at the fteenth house, on the right.
iii. If the head of HH is not home or does not agree to be interviewed, the next house on the right will be selected.
iv. Only households who have been in the camp for at least six months will be interviewed. When only one head of HH is present or in
the case of single-headed household, an enumerator of the same sex as the refugee will conduct the interview.
v. At least 10 interviews per block will be conducted.
c) Key steps for the eld data collection
i. Three different questionnaires or guides to collect primary qualitative and quantitative data will be used: One semi-structured guide
for FGDs; one semi-structured questionnaire for interviews with key informants from return communities; one questionnaire for HH
interviews. Questionnaires will include both closed and open-ended questions to obtain specic information and allow respondents to
express their views. Parts of the HH questionnaire will be taken from the questionnaire used by PRM (2011) for comparative purposes
and to enable the possibility of building on earlier results.
ii. Once approved by DRC, questionnaires will be tested with a small number of Somali refugees in Nairobi, with support from the
Refugee Consortium of Kenya.
iii. A one-day training session for enumerators will be conducted in each location. The training will include a review of the protocol,
guidance on how to conduct individual interviews and FGDs, and how to record the data.
iv. Following the training, interviews and FGD will be conducted over an established number of days in each location, with close support
and supervision from the research team.
v. A discussion on the questionnaire used for key informant interviews in communities of return will be organised with child-monitoring
partners. If necessary, support will also be provided during the data collection.
vi. The information collected will be compiled in an online database, under the supervision of the data manager. Data analysis will follow.
* It is suggested to conduct a lower number of interviews in Ifo II and Kambioos because of the smaller size of these camps compared to the other Dadaab camps, but also based on the
assumption that the population in all three locations will present similar characteristics. Most refugees in the three camps have arrived in 2011 due to the food crisis in Somalia
**Only two of the three camps (Kobe, Hilaweyn and Buramino) opened in 2011 to host people arriving due to the the food crisis are included in the planning, as it is expected that refugees in
the three camps will present a relatively similar profile
DURABLE SOLUTIONS | 55
Annex B: Interview planning
Ethiopia
Camp/Site HH Ind. %
# HH
ITW FGD*
Aw-barre (Jijiga), 2007
2 291
13 509
8,05%
20
1, 2, 6
Kebribeyah (Jijiga), 1989
2 207
16 305
9,72%
40
3, 4, 9
Sheder (Jijiga), 2008
2 610
11 497
6,85%
20
5, 7, 8
Bokolmanyo (Dolo Ado), 2009
9 831
39 363
23,47%
20
1, 2, 4, 3, 7
Melkadida (Dolo Ado), 2009
9 297
40 759
24,30%
20
5, 6, 10, 8, 11
Hilaweyn (Dolo Ado), 2011, 60 blocks
6 392
26 514
15,81%
20
7, 8, 10, 1, 2
Buramino (Dolo Ado), 2011
5 034
19 774
11,79%
20
3, 11, 4, 5, 6
Total 37 662 167 721 100% 80 Jij / 80 DA
Source: UNHCR, 31 May 2012
Kenya
Camp/Site HH Ind. %
# HH
ITW FGD*
Kakuma, 1992
9 457
46 891
9,50%
40
3, 4, 5, 6, 7,
8, 1, 2
Kambioos, 2012
2 797
13 356
2,71%
20
11, 1, 3, 4
Ifo, 1991
34 684
107 596
21,79%
40
3, 4, 6, 9, 2, 5
Ifo 2 East and West, 2011
18 590
72 500
14,69%
20
5, 2, 4, 6
Hagadera, 1992
42 488
135 169
27,38%
40
1, 2, 7, 4, 8,
10
Dagahaley, 1992
34 169
118 168
23,94%
40
5, 8, 10, 7, 9
Total
142 185
493 680
100%
160 Ddb / 40 Kak
Source: UNHCR, 18 June 2012
Total
360
62 FGD
* FGD numbers refer to the following:
(1) Youth
(2) Women
(3) Minority groups
(4) Camp leaders, incl. women
representatives and elders
(5) "Vulnerable" refugees*
(6) Incentive workers
(7) Entrepreneurs
(8) Unemployed adults
(9) Refugees who have been there for
over a decade
(10) Arrivals from 2006 and onwards
(11) Recent arrivals (2011 and onwards)
