Academic journal article Journal of Economics and Economic Education Research

The Case for Intensive Skill-Biased Technological Change

Academic journal article Journal of Economics and Economic Education Research

The Case for Intensive Skill-Biased Technological Change

Article excerpt


The skill-premium, defined as the relative wage of college to high-school graduates, has steadily and remarkably increased over the past thirty years. Roughly, the premium has risen about 2% per year implying that the relative wage rate is three times as high as it was in 1980, the year the premium began to increase. Skill biased technological change (SBTC) is generally considered to be the cause (e.g., Bound and Johnson, 1992; Autor, Katz, and Krueger, 1998; Guvenen and Kuruscu, 2010). In this case, technological advancements in the production of goods cause the relative marginal products of skilled to unskilled labor to rise. This form of SBTC has been examined in a dynamic general equilibrium framework by Heckman, Lochner, and Taber (1998) and has been found to explain the rising average skill premium reasonably well.


Recently, however, the causality of SBTC has been called into question by Card and DiNardo (2002). Their argument can be found in Figure 1. They ask, why hasn't the wage gap profile, that represents the logged skill-premium by age, increased for every age group? Presumably, because SBTC necessarily predicts an equal change in the demand for all levels of skilled labor, Card and DiNardo (2002) label the behavior of the skill premium at cohort levels a puzzle. Though their empirics raise questions, a potential explanation can be theorized when SBTC is combined with life-cycle motives. For example, it could be the case that an increase in the return to skilled labor causes those with relatively more skills (middle to older aged workers) to economize on their skill acquisition activities thereby receiving a small wage premium. The younger workers, and therefore the less skilled, intertemporally substitute into the acquisition of skills and therefore receive a larger wage premium. The total effects of the combined lower skill acquisition rates by the old and higher skill acquisition rates by the young are a steeper skill acquisition profile and a flatter wage gap profile.

The purpose of this paper is twofold. First, we construct an intergenerational model of skill acquisition for the responses of life-cycle educational expenditures from a SBTC. The theoretical analysis employs a dynamic general equilibrium overlapping generations (OLG) model of skill acquisition drawing from Heckman (1976), Auerbach and Kotlikoff (1987), Heckman et al. (1998), Fowler and Young (2004), and Guvenen and Kuruscu (2010). The model represents an extension in one important way: the unskilled do not participate in risky capital markets. This feature replicates the well-known fact that equity ownership and education attainment are highly correlated (Haliassos and Bertaut, 1995; Bertaut and Starr-McCluer, 2002) and a large percentage of the population, roughly 43.1%, never hold risky equity assets (Mankiw and Zeldes, 1991; Guiso, Haliassos, and Jappelli, 2002).

An interesting feature that results from the model's skill acquisition sector and limited participation assumption is that biased technological improvements (in favor of skilled labor) can enter in two important ways: the final goods sector and the skill acquisition sector. If the relative productivity of skilled labor increases in the final goods sector, then the demand for skills is indirectly affected. Alternatively, if the labor's productivity in the acquisition of skills increases, then the demand for skills is directly affected. Technological change occurring through the demand for labor via the production of goods is labeled extensive SBTC and through the supply of labor via skill acquisition is denoted intensive SBTC.

Until now, extensive SBTC has been the focus of study within the literature. We find that the extensive margin alone cannot account for a flatter wage gap profile. Instead, a combination of extensive and intensive SBTC is required to replicate this puzzling empirical fact. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.