Modelling geographical graduate job search using circular statistics

Theory suggests that the spatial patterns of migration flows are contingent both on individual human capital and underlying geographical structures. Here we demonstrate these features by using circular statistics in an econometric modelling framework applied to the flows of UK university graduates. While the traditional statistical techniques applied in economics do not fully account for the complexity of mobility flows, these techniques to study directionality are still relatively unexplored. The database used includes observations on around 30,000 British graduates. Our results suggest that a higher level of human capital is associated with more unidirectional flows as higher human capital migrants selectively target specific labour markets.