Alex Odlum, ACAPS
Achieving better information and evidence for decision making is essential to improved international cooperation and governance of migration. The New York Declaration for Refugees and Migrants recognises the importance of improved data collection, and calls for disaggregated data on topics such as: “regular and irregular flows, the economic impacts of migration and refugee movements, human trafficking, the needs of refugees, migrants and host communities and other issues.”
That there are gaps in the information landscape, particularly on irregular migration and smuggling, is clear. Too seldom are migration actors able to pin down the exact number of people on the move at a given time, clarify their current and future migration intentions, accurately break down demographics and highlight vulnerable groups, and identify gaps in critical service provision.
Generating new data, however, is not the only way to fill these gaps: a lot can be learned through better sharing and analysis of the information that already exists, albeit scattered across a range of actors. This blog post aims to generate further discussion on how data sharing can improve the collective understanding of migration, touching on some key issues including: the information sources available, the challenges inhibiting better analysis and some good practices.
Diversity of the information landscape on migration and displacement
The primary sources of data on admissions, readmissions, returns and reintegration are typically held within the domain of supra-national, national and sub-national authorities. Border, law enforcement and immigration agencies collect data on a wide range of metrics relating to in- and out-flows, and the legal status of those involved. Municipalities collect data on the need for health, education, transportation, sanitation, solid waste and other services. Sometimes voluntarily, and other times due to international obligations, states make some or all of these metrics public, either through their own national statistical offices, via international and regional organisations, such as the OECD and Eurostat, or through collaborative networks such as the European Migration Network.
In the context of large-scale migration and displacement flows, such as those seen from the Middle East to Europe in 2015-16, states may also delegate emergency response roles to the international community, comprised of UN agencies, the Red Cross/Red Crescent, and international NGOs. These bodies collect, organise and often disseminate vast quantities of data on arrivals and registrations, flows, people in need of assistance, and other indicators. Local civil society groups and smaller NGOs generate still more data from their own programmes, which they then use to improve their own efforts, or to lobby other actors, including local and national governments as well as INGOs.
Academics, think tanks, investigative journalists and other researchers not only analyse the above data sources, but often generate their own quantitative and qualitative datasets, both on a micro-level and on a larger scale. Furthermore, researchers often benefit from a level of neutrality and trust that allows them access to certain key stakeholders, such as smugglers and irregular migrants, that may be hesitant to interact with other, mandated, actors.
In addition, a growing body of practically focused studies aim to enrich understandings of migration and displacement, by analysing the thought processes and perspectives of people on the move, or considering moving, such as Ground Truth Solutions’ perception studies and IOM’s flows analysis. Big data analysis taps into information refugees and other migrants generate by their actions and communications, for example on social media. Used appropriately, big data can serve as a rich source of information on the movement patterns, needs and access to services of refugees and other migrants, while engaging with social media enhances two-way communication with communities.
The range of actors generating data on migration raises the question of how information can be shared in a way that enrichens our collective understanding and ability to respond, without jeopardising the safety and privacy of those involved. However, to grasp the potential for more collaborative approaches to data sharing between actors with such diverse mandates and motives, we must first recognise the challenges to deeper cooperation that need to be overcome.
What are the challenges to better data sharing?
One challenge is the lack of cooperation between states, particularly along key migration routes. While destination countries benefit from knowing more about migration flows, origin and transit countries do not always have the capacity or interest in providing more information on outflows and throughflows. For example, a lack of information sharing had inhibited earlier efforts to analyse and respond to the needs of people on the move from Turkey to Greece in 2015, although an agreement was eventually reached in early 2016 on a joint EU-Turkey data collection process. Even when destination countries invest in data capacities in transit and origin countries, they are often interested in asking questions to inform their own policies, and not necessarily those of transit countries.
Even within multinational blocs, such as the EU, data sharing on migration is far from a given. In October 2015, the European Council managed to establish an Integrated Political Crisis Response mechanism for data sharing but only after serious mistakes had been made, such as Frontex’s double counting some 710,000 migrants who entered the EU, first via Greece, and then EU countries north of the Balkans some days or weeks later.
A related challenge to more open data sharing on migration is the political sensitivity of the data. Large numbers of arrivals, for example, can be interpreted as a reason to limit further immigration. Small numbers can be cited to deprioritise assistance. The political temptation to manipulate such data is rife. Italian authorities were accused of withholding data on the number of people it rescued at sea in the months leading up to the 2016 referendum. Similarly, the Greek government has consistently maintained that around 62,000 refugees are present in Greece, in order to justify continued emergency funding from the EU, despite the humanitarian community, including UNHCR, counting some 13,000 less.
Security interests also limit data sharing, at times detrimentally. Law enforcement agencies operating within a security focused framework are not inclined to share their data to avoid jeopardising operations, which leaves potentially useful information out of reach of protection actors.
Divergent definitions and a lack of comparability further constrain the potential for more open data sharing. The parameters of blurred concepts like smuggling and trafficking can vary between countries and organisations. Even key terms, such as the definition of a migrant, is measured in multiple ways, such as country of birth, nationality, or length of stay. Although guidance is slowly emerging, a lack and capacity on what to collect, how to analyse it, and how to translate findings into useful material for decision making, leads to overlap and gaps in actionable evidence for response.
Good practices and ways forward
Despite the multiple challenges, there are also opportunities to improve data sharing and cooperation, and a growing set of good practices on which to build. International organisations and NGOs have been leading in this field, establishing spaces to facilitate data sharing and analysis. IOM, for example, has established the Global Migration Data Analysis Centre, which is compiling a global migration data portal, and providing guidance on innovative migration data analysis approaches. Humanitarian agencies are increasingly maintaining data portals, needs assessment registries and common datasets. A good example is the Humanitarian Data Exchange, which aims to build up a common repository of information available for deeper analysis of displacement and other humanitarian crises.
Other agencies are making use of their unique access to people on the move to create new, more targeted information and share it publically. 4Mi not only collect data on migration flows in hard to reach areas, but also make that data available in user-friendly platforms to encourage deeper research and analysis. IOM’s Displacement Tracking Matrix captures, processes and disseminates information about displacement and population mobility.
In addition to collecting, compiling and sharing useful data, more initiatives are prioritising analysis of the information that is already available, including ACAPS, whose analytical tools and products help make sense of information in emergencies for decision makers. Some NGOs are also combining analytical capacities to avoid duplication and fill information gaps: The Mixed Migration Platform itself is an example of NGOs combining their skills to help bridge humanitarian operations with academia and policy makers concerned about complex migration dynamics and the protection of people on the move.
Even states and their law enforcement agencies are opening up in the face of significant pressure for more transparency on migration-related data, as evidenced by Frontex making some its datasets available for public analysis. Social media companies are finding new ways to make use of their immense data to map disaster-induced displacement and migration patterns, as demonstrated by Facebook’s latest collaboration with UNICEF, IFRC, and WFP.
What are the boundaries?
The general trend towards data sharing, while necessary, does not come without risks. Some of the same challenges that hamper deeper coordination between actors with divergent interests, also explain the reluctance to cooperate more openly. Law enforcement agencies are primarily concerned with preventing and stopping irregular migration. But humanitarian organisations often see these objectives as contrary to their own goals of protecting and assisting people on the move.
While there is room for more trust between partners, particularly since states have committed to principles of data protection and privacy in the Global Compact on Migration, deeper cooperation and sharing ought to nonetheless be approached with caution. As outlined by data ethicist, Linnet Taylor, rather than producing knowledge for knowledge’s sake, knowledge production should be clearly tied to a theory of change that connects the data activity to the ultimate goal of refugee and migrant protection. So long as we bear this sage advice in mind, further steps towards data sharing and analysis should be explored, before diving into new forms of data collection.
Lastly, but perhaps most importantly, data sharing and more collaborative analysis is only an initial step towards evidence-based migration policy making. There is an onus on all those producing and analysing data to ensure findings are adequately communicated to policy makers. In turn, policy makers should do more to ensure they use findings that result from improved data sharing and analysis efforts, particularly to inform policies at the national, regional and global level.