Asset Specificity: Theoretical Considerations and Empirical Implications

Abstract

The concept of asset specificity has become very prominent in the literature on skill formation, welfare states and labour markets. While acknowledging remaining empirical obstacles in measuring asset specificity, this paper takes a more theoretical approach, arguing that a significant share of the remaining ambiguities are due to the vagueness of the underlying theoretical concept. More specifically, it is argued that the common notion of asset specificity as developed in the Varieties of Capitalism literature blurs the important distinction between the actual content of skills provided in firms and their real portability. This blurring has the unfortunate effect of detracting from the real question of interest, namely why firms invest in the formation of transferable skills. This question is approached by looking at the training decision from the perspective of firms and of workers, respectively, and by developing theoretical expectations on the impact of labour market institutions, vocational training systems and industrial relations on this decision. Subsequently, a preliminary typology of skill regimes is presented that exhibits the large variation in skill regimes to be found in advanced democracies - which stands in contrast to the simplifying dichotomy between liberal and coordinated market economies (or general and specific skill regimes) as it is purported in the VOC approach.