Abstract
In this article, we examine labour market polarization dynamics in Türkiye. First, we use highly refined microdata to classify tasks – mainly abstract, routine and manual – to conduct analysis at the occupation–task level. Second, we find evidence for polarization driven by technological changes (the routinization hypothesis), education and increasing female employment in both low- and high-wage occupations. Lastly, we analyse the tasks performed by workers in their respective occupations and find that the occupational assignment of tasks determines their value, suggesting the existence of a structural task content.