Skill-biased Technical Change: Theoretical Concepts, Empirical Problems and a Survey of the Evidence

The structure of wages and employment has shifted against the low-skilled in manyOECD countries over the last decade. Many authors have attributed this shift to theimpact of new technologies, and or technical change in general. This paper investigatesand structures the growing body of literature on skill-biased technical change (SBTC) byfirst presenting a model in which SBTC is formalised and decomposed into factor andsector biases of technical change. We show that as we go down to the job level the scopefor pure within unit-skill bias decreases and between-unit effects explain the within-uniteffects detected at higher aggregation levels. Second, we address some potential sourcesof skill bias, which are learning, R&D, human capital formation, organisational changeand the introduction of new general purpose technologies. Finally we present someconceptual and practical problems we encounter when studying SBTC empirically. Weconclude with a survey of selected empirical literature on the subject and discuss theresults in light of the empirical and theoretical problems pointed out above