Data innovation for migration: why now and how?

By Marzia Rango, Data Innovation and Capacity-Building Coordinator at the Global Migration Data Analysis Centre (GMDAC), IOM – UN Migration, and Michele Vespe, European Commission Joint Research Centre (JRC), Demography, Migration and Governance Unit, Big Data for Migration Alliance (BD4M)

Now more than ever we need to invest in responsible data innovation for the analysis of mobility and migration

The impact of COVID-19 on the production of migration statistics around the world has been severe, particularly across low- and middle-income countries. In Africa, where national population censuses and household surveys are the main sources of data on migration, travel restrictions, lockdown measures and closure of government offices have heavily affected the ability to collect data from these sources, delaying the (already infrequent) production of migration statistics. The same has occurred in some European countries. And even in countries that were able to switch to remote modalities for data collection, challenges persisted, particularly in terms of the quality of data. Meanwhile, only just over a third of the 47 African countries surveyed in May 2020 reported using sources other than traditional ones.

One of the UN Secretary General report’s (“From Promise to Action:  The Global Compact for Safe, Orderly and Regular Migration”) key recommendations is to ‘strengthen evidence-based discourse on migration.’ But how to do so when even basic facts about migration in many countries around the world are largely unknown, because capacities to collect, or properly analyse and disseminate reliable statistics are extremely modest? And when a global pandemic further limits the availability of data from traditional sources?

While ‘bold calls for better migration data are one of the oldest rituals at modern international institutions and despite the noteworthy progress over the past few decades, much remains to be done. Moreover, strengthening evidence on migration will not automatically translate into a balanced public debate on the topic, or human-rights-based, development-inducing policies, given the complex relationship between evidence and its use for policy. But all these things certainly cannot materialise without strong evidence in the first place.

The COVID-19 pandemic has increased the opportunity cost of investing in new data sources and innovative methodologies

To improve data on migration, we will need to work on several fronts simultaneously. First, substantial and well-coordinated investments are needed to enhance low- and lower-middle income countries’ capacities to regularly collect and analyse basic data on migration, at least through the inclusion of key questions on migration in censuses and surveys. Second, more efforts should be directed towards turning data routinely collected by national administrations (border data, applications for residence or work permits, etc.) into migration-relevant statistics.

Third, more investments are needed to unlock the vast and, so far, largely untapped potential of the huge amounts of data held by private companies – mobile phone operators, social media platforms, money-transfer operators, etc.. These data are already collected for commercial purposes but not yet systematically used for public policy purposes in most countries around the world. Exploiting this potential in ways that are ethically responsible and sustainable will be critical to successfully integrating these data into national statistical systems.

New data sources allow to fill some of the persistent gaps in conventional sources and methods. They can be used to get timely insights at a frequency and a level of detail that cannot be obtained through traditional sources, and on aspects of migration that are not yet fully observable, such as temporary forms of migration, aggregate and anonymised migrants’ socio-economic profiles, or how slow onset environmental change affects people’s mobility, to name a few. Just over the past five years, these data have increasingly been used to, for instance, estimate numbers of migrants in close to real time, track displacement in the aftermath of natural disasters, anticipate future movements, understand public attitudes towards migrants and refugees, monitor patterns of urban segregation and inclusion, or even sample migrants for dedicated surveys.

More recently, new data sources have helped understand and address the mobility-related impact of the COVID-19 pandemic. Anonymised and aggregated data from mobile phone operators, but also data offered by Google and Facebook, have contributed to monitoring the impact of COVID-related restrictions on mobility, as well as compliance with mobility restrictions and physical distancing measures at local and national levels, and across countries. This is demonstrated by an on-going Business-to-Government initiative between the European Commission and European Mobile Network Operators. Data innovation could help to understand how migrant communities in different countries are affected (often disproportionately) by the pandemic and inform appropriate responses.  

Looking ahead

To effectively unleash the potential of new data sources for the analysis of mobility and migration during public emergencies and beyond, data and policy communities need to be better prepared. COVID-19 has given impetus to data innovation, creating an urgent demand for timely statistics to inform effective measures and evaluate their effects. Because responses are often fragmented, ad-hoc and not fully institutionalised, they can dissipate as quickly as they were set up, with modest actual impact and huge missed opportunities.

This means more efforts should immediately be directed towards: a) developing data governance frameworks, strengthening data sharing mechanisms including “data altruism”, and creating collaboration schemes between different actors – especially between businesses and public institutions – to foster data availability; b) developing a set of ethical principles and strengthening safeguards for responsible use of private data for public policy; c) investing in new applications and cross-sectoral partnerships centred around clearly defined policy questions; and d) expanding dialogue and exchange of good practices, making knowledge about data and data innovation on migration accessible to the public.

Increasing trust in the data, from any source, and being clear about the limitations of these, should be a must-have responsibility for anyone working on data and policy on migration.