Labor Market Concentration Does Not Explain the Falling Labor Share and Other Essays in Economics

Abstract

This dissertation is comprised of three essays. The first and third chapter explore possible causes of the falling U.S. labor share. They show, first, that labor-market concentration is an implausible driver of the falling labor share and, second, that within-industry wages are not keeping up with output. The second chapter ins concerned with private equity and the financialization of the U.S. economy.

Using U.S. administrative data, Chapter 1 shows that the employment-weighted average labor market concentration has been declining since 1980 -- the opposite of the change needed to explain the falling labor share. The relationship between wages and labor market concentration has also weakened (become less negative) over that time. Together, these results make labor market concentration an implausible driver of the falling labor share despite a strong, negative relationship between labor market concentration and wages.

Chapter 2, work with Steven J. Davis, John Haltiwanger, Kyle Handley, Josh Lerner, and Javier Miranda, studies the impact of U.S. private equity buyouts on firm-level employment, job reallocation, wages, and labor productivity. Our sample covers thousands of buyouts from 1980 to 2013, which we link to Census micro data on the target firms, their establishments, and millions of comparable firms and establishments that serve as controls. Our results uncover striking differences in the real effects of buyouts, depending on the nature of the target firm, GDP growth, and credit market conditions. Employment at target firms shrinks by nearly 13% relative to controls over two years in buyouts of publicly listed firms but expands by 11% in buyouts of privately held firms. Slower GDP growth after the buyout brings lower employment growth at targets (relative to controls), as does a widening of credit spreads. Buyouts lead to productivity gains at target firms relative to controls – nine percentage points, on average, over two years post buyout. Tighter credit conditions at the time of the buyout are associated with much larger post-buyout productivity gains in target firms. A post-buyout widening of credit spreads or slowdown in GDP growth sharply curtails or reverses productivity gains in public-to-private deals.

Chapter 3 documents that the falling labor share comes entirely from decreases of within-sector labor shares while industrial-output reallocation and changes in self-employed output have counteracted some of the effect. Additionally, the decline in Manufacturing's contribution to the aggregate labor share is almost exactly offset by the increase in Services'.