Ethiopia’s jobs compact is a great opportunity for refugees. It is also a great opportunity for governments in host countries, and their international humanitarian and development partners, to learn and improve from recent experiences.

Accessing jobs for refugees is a way out of aid dependency – but only if refugees are given the opportunity to do jobs they want that will improve their lives and livelihoods.

Jordan’s experience shows that jobs compacts only work if jobs match both the skills and the aspirations of refugees. This includes non-work-related aspirations such as family unity, or access to education for their children.

This is something we heard time and time again when researching the lives and livelihoods of Syrian refugees in Jordan. In Jordan, jobs in special economic zones were unattractive to refugees due to distance, work conditions, low pay, and gender norms that prevented women taking on certain work opportunities.

Refugees also saw work permits as potentially more exploitative than conditions in the informal labour market. They tied refugees to employers, giving employers excessive power.

For Ethiopia’s jobs compact to work, the government should survey refugees’ aspirations and skills. This would help inform the type of work and therefore the type of interventions needed to support real economic opportunities for refugees. 

Our research with refugees and other migrants has shown that very little information on government policy actually filters down to the people it is meant to target. This means it is less likely to achieve the desired impact.

For example, Syrian refugees in Jordan seemed unaware of even the most basic details of the Jordan compact, including sectors covered and the fact that work permits were subsidised in the first year. This contributed to low take up.

Eritrean refugees in Ethiopia often lack accurate knowledge about the resettlement process, their eligibility status and the general likelihood of acceptance. This uncertainty keeps people waiting for months, and often years, foregoing the pursuit of alternative strategies that could make them better off in the short term.

For refugees and migrants to be able to make informed decisions about their options, information about the Ethiopia compact should be transparent and accessible. This includes information on eligibility, the conditions for participating, and potential benefits for refugees.  

It is also important to think about how to communicate this information. People use trusted information to make their decisions, drawn from their social networks. Governments and donors should communicate through organisations that have trusted relationships with refugees, including refugee organisations, community-based organisations, and influential local actors such as faith-based groups.

Ethiopia’s compact promises employment for 30,000 refugees. It is meant to create opportunities for refugees to live a normal life and reduce the pressure for onward migration, as well as contributing to Ethiopia’s economic development.

But it also has its risks.

Existing and planned industrial parks are all located along important transport routes and close to major urban centres. But refugees are currently hosted in more peripheral areas – usually along borders with neighbouring countries. To take up these jobs, refugees need to be supported to leave the camps and settle near these industrial parks.

Resettling large numbers of refugees may lead to increased tensions with the local population also hoping to access newly created jobs. Unemployment (and under-employment) and landlessness are growing problems in Ethiopia: a source of considerable discontent among young people who have invested in their education with a view of moving out of agriculture into manufacturing or service jobs. Many flock to these same rapidly growing urban centres.

Ensuring that resettled refugees have decent living standards and are protected from the risk of exploitation will also place an additional burden on already overstretched municipalities to provide housing and services including education, health, energy and transport.

Balancing the needs and aspirations of both refugees and local populations for newly-created jobs in industrial parks will need careful consideration and management. Key to this is making selection criteria for job applicants fair and transparent and employment conditions equal for both groups, while at the same time protecting the status of refugees.

The jobs compact in Ethiopia aims to increase employment opportunities available to refugees while ensuring their protection needs continue to be met: this is a challenging balance to strike.

Top-down job creation schemes like industrial parks become an attractive option, offering refugees reliable work within a controlled environment, reducing the political challenges that may be posed by greater levels of internal mobility.

But if this is the only real choice offered, there is a risk that refugees end up trapped in a situation no better than a camp, with their dependency merely transferred from the United Nations to their new employer.

If refugees have moved to a part of the country where they have fewer bonds of kinship with the surrounding populations, then they may find themselves with even fewer alternative means of support if their new job doesn’t meet all their needs.

The Ethiopian government and their international partners must therefore treat all of the government’s pledges announced at the 2016 UN summit on refugees and migrants as an interlinked package.

The provision of jobs alongside implementation of the other pledges – providing greater access to education, giving greater clarity on what local integration might look like, and creating a stronger legal framework – could create a genuine pathway for refugees to become self-reliant and confident in their rights.

But this will require policymakers to develop a more coherent approach to the different strands of reform. All parts of the Ethiopian government involved – those responsible for refugees, those developing industrial parks, those in charge of the regions hosting refugees – should come together to set out a clearer common vision.

Without this coherent vision, there is a risk that the support provided by Ethiopia’s international partners will be fragmented and fail to meet the opportunity presented for both refugees and hosting populations.

Ethiopia is rightly focusing on industrial parks to support industrialisation. Hawassa Industrial Park is a recent example, creating more than 20,000 jobs within a year. Many more parks are currently being built.

These industrial parks – Ethiopia’s term for special economic zones – are critical policy tools that can help a country transform its economy and create jobs (including potentially for refugees, directly or indirectly). But it takes time and lots of deliberate action to get them right.

SEZs that have successfully supported a country’s economic transformation can be found in China (such as Shenzhen), Singapore (Jurong), Malaysia (Penang) and Ireland (Shannon).

Africa has seen more mixed results: some zones were in weakly-connected locations; some targeted the wrong sectors; some contained very different and unconnected firms meaning firms did not benefit from spillover effects through linkages and clustering among firms. Some failed to link with the local economy, or were not supported by effective industrial strategies.

Successful zones require upfront investment in specialised infrastructure, but also, over time, complementary investment promotion activities, programmes developing links with the rest of the economy, and overall strategies focusing on skills and technological development.

Our analysis suggests that transformation succeeds only by combining such targeted support with a broader enabling environment.

Policymakers must also continually learn and adjust what’s not working. This can take time – but it’s worth remembering that it took 5-10 years before Jurong Island’s economic zone in Singapore and Shannon Free Trade Zone in Ireland generated net benefits for the country.