Heterogeneity behind skill-relatedness

Keywords: Skill-relatedness, Labor mobility, Admin data, Sweden
Session Type: Virtual Paper Abstract
Day: Sunday
Session Start / End Time: 2/27/2022 11:20 AM (Eastern Time (US & Canada)) - 2/27/2022 12:40 PM (Eastern Time (US & Canada))
Room: Virtual 34

Abstract

The existence of skill-relatedness stems from our labour market behaviour. The nature of the labour flows estimated in this way is crucial to understand in many areas, but surprisingly little is known about the geographical and socio-economic aspects of the formation of the skills network. A deep-rooted social structure underlies the creation and persistence of links between industries, occupations, or technologies. In practice, mainstream estimation methods mask such regional differences. Most research on skill linkages usually aggregate or average data across industries and regions. These simplifications, however, hide some critical information about the underlying structure of the skills that the labour flow network is trying to capture. Moreover, simplification is a waste of information and compromises the internal validity of the research by systematically under- or even over-measuring linkages for specific labour market groups or regions. In this paper, we use models from information theory to explore the geographical fragmentation and heterogeneity of Swedish labour market regions. The results show significant variance between the different types of regions; the lower the population density of a region, the more it differs from the national pattern. We also find that the differences are even more pronounced when we decompose the patterns by social and economic segments such as immigrant status, gender, and level of education. We believe our findings help calibrate policy interventions better, mainly when they target a specific region or a particular labour market group.