Cheng, Yang2024-05-302024-05-302024-05-29vt_gsexam:41003https://hdl.handle.net/10919/119173The Inflation Reduction Act has allocated over $369 billion to expedite the transition from fossil fuels to renewable energy. Along with these incentives, the funds support job training initiatives, like the recently introduced American Climate Corps. The transition to new energy forms will result in structural changes in the labor market and the demand for new and emerging skills, tasks, and jobs. A challenge, however, is that there are no existing definitions of what constitutes green jobs and skills, and thus, no clear consensus on the training workers will need for these jobs. This dissertation employs a data-driven approach using the Occupational Information Network to define and characterize green tasks, skills, and jobs. Using Natural Language Processing, we develop a method to quantify the "greenness'' of tasks and occupations. Utilizing this index, we explore the significant role of green skills during economic transitions. Our findings offer a comprehensive roadmap for understanding the evolution of green jobs and skills over the next decade. This dissertation comprises three chapters analyzing the tasks, skills, and jobs in the green economy. The first chapter investigates what constitutes green jobs and their characteristics. We construct "Task Greenness Scores" and "Occupational Green Potential" indices using Natural Language Processing and machine learning techniques to assess the greenness of tasks and overall occupations. Clustering methods categorize occupations based on task attributes -- green potential, frequency, importance, and relevance, identifying five distinct groups. This classification reveals significant variability in job greenness; although many jobs incorporate green tasks, only 113 occupations are definitively categorized as green. These are further divided into "High Green Intensity-Task Focus" and "High Green Intensity-Use Focus" groups, with the latter typically requiring less formal education and emphasizing manual skills over analytical or interactive skills. Our analysis also indicates a modest overall unconditional green wage premium of 3% for 2019 and 2020. The second chapter delineates green skills and maps their prevalence across the U.S., focusing on coal-mining communities in Appalachia. We sort a variety of skills into categories reflecting task and skill differences between green and non-green occupations, identified through O*NET. Principal Component Analysis helps categorize these into broader green skill groups such as "Technical Skills", "Management Skills", "Science Knowledge", and "Integrated Knowledge". The prevalence of green skills is notable in production-related occupations, suggesting essential technical expertise for the green economy. Interestingly, sectors traditionally viewed as energy-intensive also show a foundation conducive to green practices. Our findings highlight the necessity of tailored training programs that cater to diverse educational backgrounds, particularly emphasizing the lack of green skills in Appalachian regions, which may exacerbate inequalities during the economic transition. The third chapter examines the mediating role of green skills in local labor markets amidst the transition to a sustainable and energy-efficient economy. This chapter informs policy debates on large-scale green fiscal plans of the 2009 American Recovery and Reinvestment Act. We discover that regions well-prepared for environmental regulations or new energy development benefit from a robust stock of green skills. However, our analysis suggests that green ARRA investments are negatively correlated with wages and job creation, contrasting with positive correlations found in non-green ARRA investments. This chapter concludes that green skills significantly influence labor market outcomes, particularly in the manufacturing sector, and highlights the spillover effects of green stimulus on neighboring labor markets.ETDenIn Copyrightgreen tasksclean economyjobsskillsDissertation