Study: Firms often use automation to control certain workers’ wages
Automation Targets Higher-Earning Workers
While automation and artificial intelligence are often seen as broad threats to jobs, new research from MIT shows that U.S. companies tend to automate roles held by workers earning higher-than-average wages—a “wage premium.” This trend, observed since 1980, mainly affects non-college-educated workers who had achieved higher pay than peers, reshaping both the labor market and wage distribution.
Impact on Inequality and Productivity
The study, led by MIT economist Daron Acemoglu and Yale’s Pascual Restrepo, finds that automation is responsible for over half of the increase in income inequality from 1980 to 2016. Notably, about one-fifth of this inequality stems from firms specifically aiming to eliminate wage premiums. At the same time, the study indicates that focusing automation on cutting labor costs has muted productivity growth—offsetting 60-90% of potential productivity gains from technology.
Key Findings
- Automation often replaces workers in the 70th to 95th earning percentile within affected groups.
- The approach boosts profits by reducing wages, but doesn’t significantly enhance overall productivity.
- Income gaps between capital, highly skilled, and other workers continue to widen.
Broader Implications
Rather than maximizing tech-driven productivity, U.S. firms primarily use automation to control wages. The study suggests that recalibrating automation strategies could unlock greater productivity—benefiting both companies and workers. As Acemoglu notes, "It’s all a choice, 100 percent."
For more insights, see the original article on MIT News.