Abstract
This study reassesses Abdel Abdellaoui et al.’s findings that genetically selective migration may lead to persistent and accumulating socioeconomic and health inequalities between “types” (rich or poor) of places in the UK. Their work categorized migrants who moved to the same “type” of place (rich-to-rich or poor-to-poor) as non-migrants. We re-investigate the question of genetically selective migration by examining migration patterns between places rather than “place-types” and find genetic selectivity in whether people migrate rather than where. For example, we find evidence of positive selection of people with genetic variants correlated with better education moving from rich to poor places—a finding obscured in earlier work that used a non-standard measure of migration.
Introduction
In 2019, a high-profile study by Abdellaoui et al. (AA) proposed that the United Kingdom was undergoing a process of "genetic sorting." The hypothesis suggested that individuals with higher genetic predispositions for educational attainment and health were systematically migrating from economically deprived areas (such as former coal mining regions) to more prosperous ones. This "selective migration" was thought to exacerbate regional inequalities, leaving behind a population with lower genetic predispositions for socioeconomic success.
However, the validity of this conclusion depends heavily on how one defines a "migrant." This paper provides a critical reassessment of the AA findings, arguing that the original methodology conflated geographic stability with socioeconomic stability. By refining the definition of migration, we demonstrate that the act of moving itself—regardless of the destination—is the primary driver of genetic selection.
The Methodological Challenge: Defining the "Non-Migrant"
The core critique of the original AA study lies in its classification of "non-migrants." In the 2019 analysis, individuals who moved between two similar types of places (e.g., from one affluent area to another, or from one coal-mining town to another) were categorized as "stayers."
This paper argues that this non-standard definition obscures the "healthy migrant" effect—the well-documented phenomenon where individuals who are healthier, more educated, and more ambitious are more likely to be mobile. To address this, the researchers expanded the migration model into six distinct categories:
- Moved: From a coal-mining area to a non-coal area.
- Moved: Between two different non-coal areas (previously classified as stayers).
- Stayed: Remained in the same non-coal area.
- Moved: From a non-coal area to a coal-mining area.
- Moved: Between two different coal-mining areas (previously classified as stayers).
- Stayed: Remained in the same coal-mining area.
By separating those who actually stayed in their hometowns from those who moved within the same socioeconomic bracket, the researchers could more accurately measure the genetic traits associated with mobility.
Data and Analysis
The study utilized the UK Biobank (UKB), a comprehensive resource containing genetic and health data for roughly 500,000 individuals. The final analytical sample included 384,857 respondents of European ancestry. Genetic predispositions were measured using Polygenic Indices (PGIs)—scores that aggregate thousands of genetic variants associated with specific traits. The primary focus was on Educational Attainment (EA) and 12 health-related traits, including BMI, ADHD, and height.
Key Findings: The Mobility Premium
The results suggest that "whether" a person migrates is a much stronger predictor of their genetic profile than "where" they go.
1. The Selection of the Mobile
The most striking finding was that individuals moving from "rich" (non-coal) areas to "poor" (coal) areas possessed higher genetic predispositions for education than those who stayed in the rich areas. This contradicts the simple "sorting" narrative. It suggests that the traits required to move—cognitive ability, health, and perhaps personality traits like openness—are positively selected even when the destination is economically less advantageous.
2. The Vulnerability of Stayers
Those who remained in their birthplaces (the "true stayers") consistently showed the lowest genetic predispositions for education and health across the board. This was true for stayers in both coal and non-coal areas. The "genetic drain" on deprived areas is not just about people leaving for better opportunities; it is about the relative lack of mobility among those left behind.
3. Temporal Trends in Education
The study also identified a concerning temporal trend. While the mean EA PGI is declining across birth cohorts generally, the rate of decline is twice as steep for non-migrants (0.00276 SD per year) compared to migrants (0.00134 SD per year). This suggests that the genetic gap between mobile and sedentary populations is widening over time.
4. The Bipolar Exception
While most positive health traits were associated with migration, the genetic risk for bipolar disorder presented a unique case. Migrants moving to less deprived areas showed a higher genetic risk for bipolar disorder. This aligns with existing research suggesting a correlation between high intelligence and creativity and the genetic markers for bipolar disorder, indicating that certain types of high-achieving migration may carry specific genetic risks.
Limitations and Future Directions
While the findings are robust, the study acknowledges several limitations. The UK Biobank is known for "healthy volunteer" bias, meaning it may not perfectly represent the most deprived segments of the population. Furthermore, the focus on European ancestry—necessary for genetic homogeneity in this type of analysis—means the results cannot be automatically generalized to the UK’s diverse minority populations. Finally, the use of "birth" and "current residence" as snapshots misses the complexity of lifetime migration, such as return migration or multiple moves.
Conclusion
This reassessment shifts the focus of the "genetic sorting" debate. It is not simply that the "best and brightest" are fleeing to the cities; rather, the act of geographic mobility itself is a highly selective process. Regional inequalities are reinforced by a "mobility premium" that benefits those who move, regardless of their destination. Understanding these patterns is crucial for developing policies that address regional stagnation, as the challenges facing "left-behind" places may be rooted as much in the lack of mobility as in the direction of migration.
Summary updated