Labour Market Dynamics in the Twenty-First Century

Abstract

This thesis analyses how workers in the twenty-first century adjust to labour demand changes through mobility between occupations, employers and local labour markets. Technological change systematically replaces routine and manual tasks while complementing cognitive skills. Skill-biased technological change has diverse effects on the workforce. Growing regional employment polarization, declining employment of lowskilled workers, and rising educational wage gaps require deeper investigations in this context. In addition to medium and long-term demand trends, differences between occupations determine workers’ relative risk of experiencing short-term labour demand fluctuations. The direct negative effects of work-hour instability on income volatility, work-life imbalances and mental health are well-documented. However, the question remains of how workers’ intra-year work-hour instability affects their occupational mobility decisions. The main objective of this thesis is to investigate these topics. Chapter 2 shows that task changes within occupations are occupation-biased and how this affects the returns to tasks and the overall wage structure of the U.S. labour market. I construct a balanced occupation panel and exploit the updated O*NET ability data to derive two occupation-specific manual and cognitive task intensity measures. The decennial trend analysis shows that mainly non-routine cognitive occupations increased in cognitive intensity. Moreover, non-routine cognitive occupations show a larger decline in manual task intensity. A decomposition of the labour market by workers’ education and experience shows that cognitive-intensity-increasing task changes are more prevalent for workers with a college degree, younger and male workers. A returns analysis shows that the polarizing effects of task changes within occupations led to a substantial increase in the return to cognitive intensity between 2008 and 2017. Although every fifth worker changes their occupation every year, the mechanisms of occupational mobility are still not fully understood. Chapter 3 studies whether the detrimental effects of work-hour instability, such as income volatility and work-life imbalances, potentially influence occupational mobility decisions and whether changing occupations alleviates the work-hour fluctuations of individuals. I construct a measure of individuals’ intra-year work-hour variation using the longitudinal dimension of the monthly Current Population Survey (CPS). To observe occupational transitions, I track individuals through a balanced occupation panel of 430 occupations. The results show that workers with high work-hour fluctuations are likelier to change occupations from month to month. In the highest quartile of hour variation, the marginal effect is almost three times larger for women than men. Deeper investigations of the mechanisms behind the gender gap unveil that men who are married or have children in the household do not change occupations due to work-hour fluctuations. On the contrary, a positive and significant effect is found for women across all household compositions. A differencein- differences model shows that only workers exposed to highly fluctuating work hours sort themselves systematically into more stable occupations. To test the supply adjustments of the labour market in response to cognitive-biased task demand changes, Chapter 4 analyses population growth, employment and wage effects of workers with different educational attainment. Therefore, I divide the U.S. labour market into local labour markets using data from the American Community Survey (ACS). The segmentation of the U.S. labour market allows me to use the local industrial specialization for instrumenting my technological change measure based on occupations’ task demands. The causal effects show a relative increase in the population of both college and non-college workers in local labour markets with higher exposure to cognitive-biased technological change. Cognitive-biased technological change has detrimental wage and employment effects on non-college workers, including lower employment shares, reduced wages and higher labour force non-participation rates. Moreover, the downward pressure on wages of high-school workers on high-school dropouts increases the college wage premium in regions with more substantial growth in cognitivebiased task demand. This thesis contributes to the empirical labour and macroeconomic literature by unveiling new findings on heterogeneous task demand changes, work-hour instability and labour mobility. The novel results can help policymakers combat precarious working conditions, rising educational wage gaps, and population polarization between local labour markets.