Skip to content
Livelihoods programming and its influence on secondary migration

In search of greater freedoms and opportunities, thousands of Eritreans have fled their country in recent years – many directly across the border to Ethiopia. One estimate by the UN Refugee Agency, UNHCR, in 2016 put the number of registered Eritrean refugees residing in Ethiopia at nearly 800,000. But for many the journey does not stop there. Secondary migration of Eritreans from Ethiopia is considerable, with some evidence suggesting that as much as two thirds of the Eritrean population in Ethiopia moved onwards in 2015.[1]

Humanitarian organisations have long been delivering livelihoods programming in Ethiopia’s refugee camps, providing capital to help participants establish micro-enterprises or equipping them with vocational skills in a particular sector, such as tailoring or computing, and/or basic educational skills such as numeracy. Although such interventions attempt to achieve humanitarian and developmental outcomes (by generating improvements in participants’ well-being), they are also increasingly justified in the name of migration control – and more specifically the prevention of onward movement. Through 63 interviews with Eritreans across three different sites in Ethiopiawe examined the extent to which these measures actually influence the way people plan for their futures.[2]  

The logics and limits of livelihoods programming
Livelihoods programming is an example of a policy measure designed to steer migration decision making at the individual level. Part of the objective of livelihoods programming is to prevent movement before it occurs, based on the assumption that by providing greater economic opportunity and security to people living in ‘sending’ countries, it is possible to create a viable alternative to (primarily irregular) migration. It is hoped that the option of migration is then seen as relatively less attractive.

With the caveat that our research looked only at a sample of certain types of programming (mainly lending initiatives and vocational training), our evidence suggests that while such interventions are capable of helping people to meet basic needs, there appears to be little meaningful effect on more transformative change in people’s livelihoods or migration plans.

A number of respondents in Adi Harush camp in the north of the country described how they had received financial support from non-governmental organisations (NGOs), which they used to start up their own micro-enterprises such as small shops within the camp or buying and raising livestock. Many welcomed this support: one man, for example, was now running a successful horse and carriage transport service, having received an initial loan; a single mother told us that her NGO-funded chicken farm was doing well.

However, many interviewees also discussed the limitations of these programmes. Although one woman had used her loan to establish a small shop within the camp, her main problem was one of marketplace saturation, which made it hard to make a profit and expand. The point here is that although camps form their own economies, it is evident that the markets within them are often limited, creating little opportunity to accumulate reasonable profits.

We observed a similar pattern for skills and vocational training: while respondents often expressed gratitude for, and satisfaction with, the experience, they reported little significant or long-term impact. Our research points to refugees being unable to put newly acquired skills to use in their wider environment, primarily as a result of existing legislation. One man described how he had planned to get a driving licence and start working as a taxi driver, before hearing from the Administration for Refugees and Returnee Affairs and from fellow Eritreans that he – as a refugee barred from engaging in the formal economy – would not be allowed to do this. For others, the range of training programmes currently available are simply not relevant to their interests and aspirations, which deters take-up altogether.

Ultimately, these forms of livelihoods support are addressing the symptoms rather than the underlying structures of poverty and economic marginalisation facing Eritrean refugees – which are likely tobe driving the desire among so many for secondary migration. In particular, these underlying factors include refugees’ lack of rights to formal, better paid and higher skilled employment. One woman told us: “In the Netherlands you can get a job with better wages. My husband tells me that it is also much better to access the basic infrastructure services in the Netherlands than in Ethiopia. […] I know it is possible to travel to the Netherlands.” Another man illustrated what might prevent him making a similar onward move: “If I am not working, I will run out of patience and I may attempt the dangerous move to Europe but I hope I can get a job and will not need to risk this.” Without meaningful improvements in refugees’ access to decent work – work that is reliable, adequately paid, and makes use of their skills – the support provided by livelihoods programming simply cannot be expected to reduce secondary migration in any significant sense.

Improving livelihoods programming
To better align livelihoods interventions with both the intended beneficiaries’ needs and the dynamics of the local context, efforts should be made to (re)evaluate the local relevance of the training that is offered and to expand the types of work in which refugees can participate. Further to this, more could be done to provide ongoing mentoring and supervision for those refugees who have received loans and training, to help them maximise their potential gains. At the same time, however, NGO programme staff told us that administering effective long-term programming is difficult precisely because of the nature of the implementing environment. In a context of high secondary migration, interventions are intentionally designed to be quick and simple, purely to encourage enrolment; this limits innovation space and encourages inflexibility, thus placing obvious constraints on those responsible for design and implementation.

Putting questions of programming to one side, refugees’ ability to build successful, dignified lives will largely hinge on their capacity to access decent work. This is fundamental, underpinning both the challenge of survival in Ethiopia, as well as the resolve (formed by many) to move on. Although there are no guarantees that livelihood security automatically follows from moving into the formal labour market (particularly in urban contexts of generalised high unemployment), our findings suggest that people may become more inclined to remain as a result of improved access to better forms of work.  

Government steps

Recent policy developments suggest that Ethiopia is shifting its stance on the right to work. In September 2016, the government announced nine pledges to improve the rights of, and services available to, refugees in the country. These pledges have now been developed into a national ‘roadmap’ with six thematic priorities; one of these is work and livelihoods, which will include work permits for refugees, an increase in livelihoods programming and the creation of industrial parks. These parks are a key element of the Ethiopian Jobs Compact which was officially rolled out in December 2017 – a US$500 million agreement between the Ethiopian government and external donors to create 100,000 new jobs (roughly one-third of which will be allocated to refugees).[3] Such initiatives, already implemented in some major refugee-hosting countries such as Jordan, are designed to improve access to work for both refugees and members of the host community.

Based on our research, this seems a positive step in the right direction. However, much of the success of both the roadmap and the compact will ultimately depend on the detail of their design and implementation. To that end, we offer five suggestions to policymakers:

First, information about any initiative must be transparent, clear and accessible in order for refugees to make informed decisions about their options. Not only should potential participants be made fully aware of eligibility criteria but they should also be provided with information about what to expect regarding the terms, nature and returns of the work on offer.

Second, some refugees may reject jobs in industrial parks because the jobs either do not align with their backgrounds and interests or for reasons of location: recent analysis of work provided in special economic zones in Jordan, for example, shows that low take-up among Syrian refugees is often related to factory location, with people reluctant to take on long journeys or be separated from family.[4] Take-up will also be influenced by the way in which these jobs are perceived by those at whom they are targeted: are they seen to offer ‘decent’ work, or is the activity considered demeaning or exploitative?

Third, the politics of job allocation must be anticipated and taken into account. Urban unemployment in Ethiopia remains high across all groups in society, and economic growth has slowed in recent years. In a context where significant numbers of Ethiopians are also affected by poverty and vulnerability, the way in which allocation procedures are framed, conveyed and implemented is likely to affect the nature of social relations between ‘host’ and ‘refugee’ communities.

Fourth, refugee policies should be consistent with the realities of local labour markets. With widespread informal employment in Ethiopia, attempts to formalise refugee employment might be challenging. The feasibility and future sustainability of industrial parks thus need to be carefully considered in light of the country’s broader economic environment.

Finally, rather than focusing exclusively on quotas and number of work permits issued, pledges and related monitoring mechanisms should consider the extent to which policies and interventions – such as the Jobs Compact – are actually improving the lives and livelihoods of refugees and hosts.


Richard Mallett
Research Associate

Jessica Hagen-Zanker
Senior Research Fellow

Clare Cummings
Research Fellow

Overseas Development Institute 

Nassim Majidi

Samuel Hall

[1] Amnesty International (2016) Tackling the global refugee crisis: from shirking to sharing responsibility

[2] Mallett R, Hagen-Zanker J, Majidi N, Cummings C with Sturge G, Schaefer K and Vidal P (2017) Journeys on hold: How policy influences the migration decisions of Eritreans in Ethiopia ODI Working Paper 506

[3] World Bank (2017) Program-for-results information document (PID) concept stage

[4] Barbelet V, Hagen-Zanker J and Mansour-Ille D (2018) The Jordan Compact: lessons learnt and implications for future refugee compacts

This site is registered on as a development site. Switch to a production site key to remove this banner.