Assessing the quality of data on international migration flows in Europe: The case of undercounting

Undercounting is a critical issue in migration statistics, resulting in bias. This typically arises from insufficient reporting requirements and problems with enforcing such requirements. The main sources of information on undercounting are the metadata accompanying official statistics and expert opinions. However, metadata and arbitrary expert opinions may be limited by overlooking important details in migration data shared by various countries. This includes potential oversight of changes in methodologies, definitions, or retrospective updates to the data following censuses. This work presents a methodological solution with three objectives to address undercounting in international migration data. First, we provide an overview of available metadata and expert opinions on undercounting in European migration flows. Secondly, we propose a novel data-driven approach that incorporates year-specific and duration-of-stay-adjusted classifications. The proposed methodological solution relies on comparisons of flows in the same direction reported by a given country with high-quality data reported by another set of countries. We use bilateral migration data provided by Eurostat, United Nations, and selected national statistical institutes. Duration-of-stay correction coefficients are derived through an optimization model or borrowed from the literature. Metadata and expert opinion scores can also be integrated to classify undercounting. Finally, we provide a dynamic classification of undercounting for thirty-two European countries (2002-21), accessible through an online Shiny application, offering flexibility and adaptability. The findings highlight significant undercounting in new European Union Member States, particularly Bulgaria, Latvia, and Romania. Interestingly, other European countries, including those presumed to maintain reliable population statistics, also exhibit notable periods of undercounting.