Antitrust & Competition Policy Blog

Editor: D. Daniel Sokol
University of Florida
Levin College of Law

Friday, January 18, 2019

Labor Market Power

David Berger (Northwestern University); Kyle Herkenhoff (University of Minnesota); Simon Mongey (Federal Reserve Bank of Minneapolis) identify Labor Market Power.

Abstract: As U.S. Senator Corey A. Booker wrote in an open letter to the acting Attorney General and Acting Chairman of the FTC, ``A growing body of evidence suggests that the rise of concentration across U.S. industries has helped create a labor market in which fewer workers are able to fairly bargain with their employers to set their wages on competitive terms.''\footnote{See the article and link to Senator Booker's letter here: \citet{booker2017market}, \url{}} Similarly, \citet{furman2016beyond} calls for additional research on the affects of labor market concentration on wages and job flows. \textbf{Objective:} The aim of our project is to measure the impact of labor market power amongst employers on employment, wages, investment, and labor market flows. \textbf{Contribution:} Recent studies have focused almost exclusively on the role of product market power as measured by national sales concentration (e.g. \citet{gutierrez2016investment}, \citet{autor2017fall} and \citet{de2017rise}). Studies that have thought about input market power---a more local phenomenon when inputs are non-traded, such as labor---have focused on relatively narrow industries (for a summary of the recent theory see \citet{manning2003monopsony} and for a summary of recent empirics see Section 4.2 in \citet{alan2011imperfect}). Our contribution is to develop an identification strategy which isolates the affects of labor market power (e.g. \textit{monopsony} and \textit{oligopsony}) on employment, wages, investment, and labor market flows for a broad range of industries. We measure the impact of labor market power on wages using a novel identification strategy. The insight is to isolate firms that are in tradable sectors (e.g. textiles, manufacturing etc.) and then use within-firm variation in market power across regions after a merger. For measurement and identification of mergers, wages, and job flows, we plan to use the LEHD and LBD Census micro-data (we currently have active RDC projects with access to both datasets). By focusing on tradable sectors, we remove the role of product market power. Prices are set outside of the local labor market, and so a merger will not alter prices or product market power. However, the merger will change the number of employers who are hiring within a given region, thus altering labor market power. This allows us to pinpoint the impact of monopsony and oligopsony on the real economy. We develop a model of oligopsony in the labor market, adapting the \citet{atkeson2008pricing} model of imperfect product market competition to a general equilibrium labor market. We model a finite number of firms that hire within a given industry in a given region. The model yields several testable implications: \begin{enumerate} \item The wage bill and employment decline as labor markets become more concentrated. \item Pass-through to wages from idiosyncratic shocks to the firm decrease as labor markets become more concentrated.

| Permalink


Post a comment