Association for Public Policy Analysis & Management

Technological advancements have historically been important forces in reshaping society and the way people structure their worlds. Recent developments in automation and artificial intelligence promise rapid and possibly unprecedented levels of change, with impacts ranging from changes in basic routine activities to deeper changes in the nature of employment. Whereas the use of such technologies can be directly observed, other subjective dimensions and consequences are harder to capture. Yet these dimensions are equally important to drawing a fuller understanding of the consequences of technological change on labor.

This paper sheds light on these subjective matters by examining how workers perceive technological change and to what degree it affects their well-being at the workplace. To address the first set questions, we rely on a 2017 Pew Research Center survey on Automation in Everyday Life to unpack how technology affects individuals’ working lives as well as their broader perceptions of automation and expectations and concerns regarding the future. We rely on statistical analysis to estimate differential effects for workers across industries, job roles (manual labor, management, etc.), and other characteristics (age, race, gender, etc.).

Second, we use the 2010, 2012 and 2013 American Time Use Survey (ATUS) Well-Being Module to ascertain whether workers in highly automated jobs experience different levels of well-being (measured in by scales of sadness, happiness, stress, tiredness, meaningfulness) when performing their jobs. While we cannot identify precisely the degree of automation in each worker job, we use the methodology of Frey and Osborne (2017) to classify workers by their occupation’s susceptibility to computerization, leveraging detailed occupation data from the Current Population Survey and O*NET. As an alternative approach, we use the Skill-Biased Technological Change literature (Acemoglu and Autor 2010; Jaimovich and Siu 2014) to classify workers by occupation task content (degree of routinization and cognition). These alternative classifications serve as robustness checks on our findings and additionally help to assess the reliability of the occupation classification approaches themselves.

In view of impending technological change due to automation and artificial intelligence, this research sheds light on emerging questions about technological complementarity in the workplace. By expanding the scope of discourse to address the critical issue of worker well-being, we provide a more robust understanding of the impacts of technology on the lives of workers.