Migration Age Patterns and Life-Course Transitions: A Cross-National Comparison

Abstract

Internal migration is integral to human development and is the main mechanism to redistribute populations within national borders, so understanding the dynamics of human mobility is essential. Young adults are the most mobile as they move to take up educational, employment and partnering opportunities. Despite the persistent age-selectivity of human mobility, there is emerging evidence of significant cross-national variations in the age structure of migration. However, the extent of these differences is yet to be comprehensively established and the factors underpinning these differences remain poorly understood. The lack of comparative research can be traced back to (1) the lack of reliable comparative metrics, (2) the limited availability of migration data, which has favoured research primarily on North America and Europe at the expense of other regions, and (3) the absence of a theoretical framework that effectively conceptualises the diversity of migration age patterns. In this context, this thesis seeks to advance understanding of cross-national variations in migration age patterns by (1) establishing robust comparative age profile metrics, (2) systematically identifying differences in the age profile of migration across a global sample of countries, and (3) establishing the underlying determinants of migration age patterns and identifying the mechanisms through which these factors operate. This thesis uses a unique database, the Integrated Public Use Microdata Series, to assemble a novel migration dataset of 27 countries, disaggregated by single years of age. By extending the geographic scope of comparative research to all major world regions, this research uncovers the migration age patterns of under-researched regions and therefore provides a comprehensive picture of global variations. To enable robust comparative analysis, the thesis validates and adopts an innovative non-parametric smoothing technique, and establishes, through correlation and factor analysis, a novel suite of comparative metrics that capture the key features of migration age profiles and underscore differences among countries. Building on these methodological advancements, the thesis ascertains the extent of global variation in migration age profiles and identifies major migration clusters around the world. It establishes distinctive migration age patterns among the countries of South-East Asia, which are characterised by relatively young age profiles with high migration peaks. These stand in clear contrast to the lower-intensity profiles of the more developed countries, which have peaks at older ages. Latin America, on the other hand, falls into two distinct clusters differentiated by their age at peak migration. To explain these differences, the thesis proposes a proximate determinants framework, which situates life-course transitions as intermediaries between migration age patterns and the broader socio-economic context. The thesis quantitatively establishes the link between the age structure of migration and that of the life-course across the 27 case-study countries, with a focus on education completion, labour market entry union formation and first childbearing. The analysis reveals substantial diversity in the timing and spread of transitions to adult roles, and shows that cross-national differences in the age profile of migration closely parallel variations in the age structure of the life-course for over two-thirds of countries. Migration age patterns are found to align with transitions to adulthood for both sexes, but most strongly among women. Using event-history analysis for two developed nations, the thesis further elaborates the nature of these links and shows how life-course transitions combine to shape migration. Differences in the relative importance of life-course transitions underpinning migration are traced to the wider socio-economic context, which shapes the structure of the life-course and influences its interaction with migration. The thesis sits at the intersection of two well-established research traditions in demography: the life-course approach and cross-national comparative analysis. By coupling these perspectives, this analysis reveals the diversity of migration patterns and processes across the world, and provides new insights into the extent and causes of migration. It demonstrates that cultural settings, social norms and practices, economic conditions and local contingencies influence the structure of the life-course, which strongly shapes migration age patterns. The proposed framework suggests that countries follow different trajectories in the evolution of their migration age patterns, which reflect unique contextual settings.