The goal of this paper is two-fold. First, we reexamine the evidence for the capital-skill complementarity (CSC) and the skill-biased technological change (SBTC) hypotheses at the sectoral level in the US economy for the period 1970–2005. Second, we quantify their effect on the evolution of the wage skill premium. To do so, we estimate a translog model with three production factors (skilled labor, unskilled labor, and capital) for different sets of industry aggregates suggested by the literature. At the aggregated level, we find that both CSC and SBTC explain a substantial part of the observed change in the skill premium. The CSC hypothesis also receives support across sectors, although SBTC often explains a larger part of the premium change. We also find that the relevance of CSC increases with the level of aggregation of the data. Besides, when we disaggregate capital into ICT and non-ICT, our results suggest that often ICT capital is not the primary source of CSC. However, ICT-CSC is the most important driver of the skill premium in specific sectors, such as financial and business services.

本文的目标是双重的。首先，我们在1970-2005年期间重新审查了美国经济部门层面的资本技能互补性（CSC）和技术偏向技术变革（SBTC）假设的证据。其次，我们量化它们对工资技能溢价演变的影响。为此，我们估计了针对文献所建议的不同行业集合的具有三个生产要素（熟练劳动力，非熟练劳动力和资本）的跨模型。从总体上看，我们发现CSC和SBTC都能解释所观察到的技能溢价变化的很大一部分。尽管SBTC经常解释保费变动的很大一部分，但CSC假设也得到了各个部门的支持。我们还发现，CSC的相关性随数据聚合程度的增加而增加。除了，当我们将资本分解为ICT和非ICT时，我们的结果表明，ICT资本通常不是CSC的主要来源。但是，ICT-CSC是金融和商业服务等特定领域技能溢价的最重要驱动力。