December 2024
José Azar, Ioana Marinescu
Abstract:
Labor economics often assumes that wages w are equal to the marginal revenue product of labor MRPL. However, recent literature has shown that firms’ market power allows them to pay wages substantially below marginal productivity.
The markdown (MRPL − w)/w is our preferred measure of firms’ monopsony power, and captures the percent wage increase that would occur if monopsony power were eliminated. We derive the markdown across three classes of models, each embodying a distinct source of monopsony power. First, in oligopsony models, monopsony power arises from strategic interactions between large firms, and is related to labor market concentration. Second, in job differentiation models, monopsony power arises from workers’ heterogeneous preferences over jobs that differ in wages and amenities. Finally, in search and matching models, it arises from frictions that prevent workers from accessing all existing job vacancies. To identify the markdown, empirical studies often rely on estimating the firm-level labor supply elasticity and taking its inverse as a measure of the markdown. A few studies directly estimate MRPL using a production function approach. Across studies, the markdown typically ranges between 15% and 50% implying that wages would increase by 15 to 50% if firms’ monopsony power were eliminated. Finally, we analyze the policy implications of monopsony power in three areas, drawing on both theory and empirical analysis: merger control in antitrust policy, the regulation of non-competition agreements, and minimum wages. Monopsony power helps explain how mergers and noncompetition agreements can lower wages, and how minimum wages can increase employment. Overall, the literature shows that monopsony power is significant, and should be considered when analyzing policy and the sources of wage variation.
Download this paper here: https://www.rfberlin.com/wp-content/uploads/2024/12/24031.pdf
