The Effects of Skill-Biased Technical Change on Income Distributions

This thesis broadly investigates the impacts of skill-biased technical change (SBTC) on income distributions.

In Chapter 1, I develop a SBTC model with an intergenerational framework, where heterogeneously-skilled households make intergenerational transfers that determine the skill outcome of the next generation. By increasing the skill-premium, an SBTC shock increases these transfers, thus improving the probability that children from both high and low-skilled households will become skilled. With diminishing returns on investments, the relative improvement in skill outcomes is larger for children from low-skilled households compared to children from high-skilled households, implying that relative skill mobility also improves.

I test the model's predictions in Chapter 2, using data from Chetty et al. (2014) which show how college attendance rates of children in U.S. commuting zones (CZs) are linked to the rank of their families in the national income distribution. A technology measure is constructed for each CZ using its share of STEM (Science, Technology, Engineering, and Mathematics) workers, which I instrument using a Bartik-type IV to deal with endogeneity concerns. From 2SLS estimations, I find that college attendance rates of children from households in the same income rank improve if households are located in higher technology CZs, with the improvement being larger among lower-ranked households. The findings from Chapters 1 and 2 thus suggest that SBTC can improve both absolute and relative skill mobility.

Chapter 3 examines how the skill-premium from the SBTC model can impact between-group inequality among skilled and unskilled workers. I construct a Gini measure for the SBTC model, and find that if the proportion of skilled workers is more than half, a rising skill-premium can actually reduce between-group inequality if the skill-share is also growing. For empirical validation, I observe trends in the skill-portion, skill-premium and two inequality indices (Gini & Theil) for full-time, full-year workers in the U.S from 1980-2018. With the proportion of college-equivalent workers continually increasing, I find that the relationship between the skill-premium and between-group inequality has changed: from 1980-2005, the relationship was strongly positive, while from 2005 onward, it has become negative. Thus, the skill-premium's impact on between-group inequality is not unidirectional