Jobquality Determinants of Relative Employment Growth

In this paper I investigate developments in relative employ-ment in relation to general jobquality, to ascertain the validity of the polarization hypothesis, introduced by Goos and Manning (2007), in a compressed wage regime. There is a general agree-ment that Skill Biased Technological Change has not a¤ected labor market outcomes in the same magnitude in the Scandina-vian countries as it has in comparable economies. While this may be true with respect to wages and unemployment, relative employment changes (of occupations) were remarkably similar to the US and the UK. While the share of low educated workers in the employed population dropped considerably, occupations that were formerly intensive in low skilled work grew in relative im-portance. Analogously, occupations formerly intensive in medium skilled workers lost in importance despite seemingly unchanged employment opportunities for medium skilled workers. By link-ing employment growth and task content of occupations, I show that initial task content is well able to explain changes in relative employment, with a high initial routine task content being a good predictor for joboutow and a high nonroutine, interactive and manual task content bein associated with positive job growth