Policymakers and researchers seeking to understand how migration is changing globally have long faced a fundamental challenge: while data showing where migrants live are available for over half a century, there has been no comprehensive record of how migration movements between countries have evolved year by year over this period.
The challenge is becoming increasingly urgent as countries respond to demographic change, labour shortages, displacement and other forces shaping human mobility. To help address it, the International Organization for Migration (IOM), with support from the German Government and research partners, has published FLOW, the first global dataset providing annual estimates of migration movements between all countries in the world since 1960.
"Good migration policy starts with good data," said IOM Director General Amy Pope. "By showing how migration has changed over time, FLOW can help us plan ahead and shape policies that better respond to the needs of migrants and communities."
Published in Nature Scientific Data , the open-access resource provides the most comprehensive historical picture to date of international migration patterns across more than six decades. The launch coincides with Director General Pope's visit to Berlin for High-Level Consultations with the Government of Germany and reflects a shared commitment to enhancing data and insights for migration governance.
"This global new dataset demonstrates the value of cooperation between Germany and IOM," said Niels Annen, State Secretary at Germany's Federal Ministry for Economic Cooperation and Development (BMZ). "It helps us see the long history of global migration more clearly, and enables deeper analysis to inform more impactful policies and programmes in countries of origin and destination."
The dataset provides annual, sex-disaggregated estimates of migration flows between all countries from 1960 onward, significantly expanding previous efforts by offering both global coverage and year-by-year estimates over a 65-year period. It offers a new resource for understanding how migration systems have evolved and how they may change in the future.
To develop FLOW, researchers combined international migrant stock data from the United Nations and the World Bank with demographic information and observed migration flows. Using statistical modelling and machine-learning approaches, they reconstructed annual migration movements that can be compared consistently across countries and over time.
"Estimating historical migration flows is challenging because each data source only provides part of the picture," said Dr. Robert Beyer, data scientist at the IOM Global Data Institute and lead author of the study. "While uncertainties are unavoidable, the approach aims to extract as much information as possible from the available evidence to build a consistent picture of international migration over time."
The dataset is expected to support research, migration forecasting, demographic and labour market analysis, and policymaking, especially in countries where direct migration records remain limited or unavailable. The FLOW dataset and accompanying methodology are publicly available through Nature Scientific Data.
