The United Nations International Migrant is one of the premier datasets presenting global snapshots of dyadic migrant populations. The dataset records host country, stock country, and population size for 231 international countries and territories. Additionally, the United Nations provides dissaggregated datasets featuring populations broken down by age and sex. In contrast to the United Nations Populations of Concern dataset, there are a limited number of competing datasets. These include The World Bank Group’s Bilateral Migrant Stock Database,¹ and The Organisation for Economic Co-operation and Development’s (OECD) database of immigrants and expatriates.²

Figure 1: Top 5 migrant stock populations in the United States between 1990–2019.

As one of the few sources of international migrant stock data it’s regularly featured in peer reviewed research and serves as a foundation for migration policy. Several recent manuscripts from Guy Abel that incorporate the migrant stock dataset have gained notoriety. Abel and Sander’s³ study Quantifying Global International Migration Flows utilized the migrant stock database to infer dyadic flows between countries. The dataset was recently the subject of an examination of female international migration by.⁴ The 2005 revision of the migrant stock database was featured in an analysis of social networks and migrant mobilization in Southern Europe.⁵

Although the migrant stock dataset is featured in numerous peer reviewed manuscripts and commissioned reports, it still suffers from persistent issues that plague many human geography and migration datasets. There may be underlying discrepancies between countries, due to varying reporting methodologies. These may include reporting intervals, legal definitions of migrant status, or an administration’s capacity to accurately report populations. Moreover, as a result of the sensitivity of migrant populations and individual status, population estimates are often under-reported.⁶