Labor Market Power and Rent-Sharing
Paper Session
Sunday, Jan. 5, 2025 10:15 AM - 12:15 PM (PST)
- Chair: Hyunseob Kim, Federal Reserve Bank of Chicago
The Labor Market Impact of Shareholder Power: Worker-Level Evidence
Abstract
Using worker-level data from the US Census Bureau’s LEHD program from 1993 through 2015, we show that shareholder power leads to large earnings losses for employees. We track the earnings of employees up to five years after their firms experience a material increase in ownership by concentrated institutional shareholders, relative to employees of other firms that experience a similarly sized increase in ownership by diffused institutional shareholders. We find that over the next six years, the cumulative earnings of the affected employees decline by 10% of their pre-event annual earnings on average. Treated workers with earnings in the top within-firm tercile and managers (such as chief executives) experience larger earnings losses—their cumulative earnings decline by 16% and 43% relative to their respective pre-event annual earnings. In contrast, the increases ownership concentration little affect the earnings of employees in the bottom tercile. The earnings losses are concentrated among firm stayers. The collection of evidence is consistent with rising shareholder power reducing employee rents.Employer Concentration and Outside Options
Abstract
We find that increases in employer concentration causally reduce wages, using a newinstrument for employer concentration based on changes in large firms' national hiring
patterns. We also show that measuring employer concentration within a single local
occupation excludes important parts of workers' true labor markets. Moving from the
median to the 95th percentile of employer concentration as experienced by workers
causally reduces wages by 10.7 log points in low-outward-mobility occupations like
registered nurses or security guards, and by 3 log points in high-outward-mobility
occupations like bank tellers or counter attendants. We propose a new approach for
defining mobility-adjusted labor markets, measuring employer concentration on
clusters of local occupations identified through asymmetric mobility patterns (using
new, highly granular data on occupational mobility from 16 million resumes). Overall,
we estimate that around one in six U.S. workers face wage suppression of 2% or more
as a result of employer concentration.
Right to Work in the Era of Generative Artificial Intelligence
Abstract
This paper examines how U.S. labor policies and institutions, particularly Right-to-Work (RTW) laws relate to unionization and labor outcomes in the context of rapidly advancing AI, especially generative AI, which is increasingly performing tasks done in "white-collar" jobs which are historically less unionized than "blue-collar" jobs. We combine prompt engineering with traditional econometric analysis to investigate how the introduction of generative AI, specifically ChatGPT, impacts unionization and labor outcomes across occupations with varying AI exposure levels in both RTW and non-RTW states. To measure AI exposure, we develop an "Occupational Gen AI Exposure Score" by analyzing task data from the Bureau of Labor Statistics’ O*Net dataset and querying large language models like OpenAI’s ChatGPT and Meta’s Llama 3.1. This score estimates the percentage of tasks within each occupation that could potentially be performed by state-of-the-art generative AI models. With the Occupational Gen AI Exposure Scores at hand, we conduct our econometric analysis which draws on the outgoing rotation group of the monthly CPS data from January 2020 to August 2024. We employ a difference-in-difference framework to compare the impact of Generative AI on unionization and labor market outcomes in RTW and non-RTW states. We also use a triple difference analysis to examine occupation-specific responses to AI model releases. The difference in difference analysis shows that the release of ChatGPT decreased unionization rates in occupations in RTW states relative to non-RTW states. The triple difference estimate shows that the change is significant based on RTW laws. The reduction in unionization rates in Right-to-Work (RTW) states following the release of ChatGPT suggests that RTW laws may amplify the impact of generative AI on labor dynamics, particularly by weakening collective bargaining structures in occupations highly exposed to AI. In contrast, the null effect in unionization observed in non-RTW states among workers in high-AI-exposureDiscussant(s)
Andrew Garin
,
Carnegie Mellon University
Alex He
,
University of Maryland
Sydnee Caldwell
,
University of California-Berkeley
JEL Classifications
- J3 - Wages, Compensation, and Labor Costs