Labor's share or the proportion of a country's gross domestic product (GDP) allocated to labor has been attracting much attention as a subject of economic research in recent years. In advanced economies such as the United States, Japan, and Europe, the labor share has been on a declining trend for years (Figure). Meanwhile, when it comes to the reasons and background behind the decline, there seem to be 100 schools of thought. In what follows, I would like to outline some of those arguments with a particular focus on findings from recent studies.
"Real Culprit" behind the Fall of Labor’s Share
TSURU Kotaro Faculty Fellow, RIETI
♦ ♦ ♦
To begin with, we need to remember that the labor share shows different trends depending on the definition of the term and the indicator used. The treatment of labor income attributed to owner employees or family employees is one example. The decline of the labor share in the United States and elsewhere has been subject to extensive analysis lately, and there have been some studies focusing on measurement issues such as the treatment of capital depreciation, income earned by self-employed individuals, and intangible capital. However, those factors alone cannot fully explain the observed decline of the labor share.
Meanwhile, researchers have pointed to the following three economic factors as contributing to the fall of the labor share mainly in the United States.
The first factor is a relative decline in the cost of capital, which has occurred against the backdrop of a rapid fall in the (quality-adjusted) prices of information and communications technology (ICT) devices, as highlighted in a paper authored by University of Minnesota Associate Professor Loukas Karabarbounis et al. In this case, when the degree to which capital substitutes labor is large relative to changes in factor prices (i.e., if the capital-labor elasticity of substitution exceeds one), the labor share declines.
The second factor is the impact of trade and international outsourcing, as emphasized by University of Edinburgh Professor Michael W. L. Elsby et al. They showed that the labor share in the United States declined more rapidly in sectors affected by increased exports from China and other labor-abundant countries. The third factor is changes in social norm and labor market institutions, such as a lower unionization rate and a decline in the real minimum wage level.
Recently, Massachusetts Institute of Technology Professor David Autor et al. proposed a new factor to explain the fall of the labor share. They cite the rise of "superstar firms" such as Facebook and Amazon.com as the most important factor causing the global decline of the labor share. Assuming that superstar firms are those characterized by high profits and a low share of labor, as they increase their dominance in their respective sectors, market concentration will rise and the aggregate labor share will fall.
♦ ♦ ♦
Autor et al. countered the aforementioned three major hypotheses as follows.
Regarding the first factor, their analysis based on firm-level data in the United States found that the fall in the aggregate labor share is attributable to an increase in the market share of leading firms that are likely to have a low share of labor, rather than reflecting a general fall in the labor share of firms across the board. If a relative decline in the cost of capital is the primary factor, the labor share should fall in many firms as they all benefit from the lower cost of capital.
Regarding the second factor, Auto et al. concluded that there is no evidence showing that the labor share fell more rapidly in manufacturing industries with greater exposure to exogenous trade shocks than in other manufacturing industries in the United States. Noting that non-trade sectors—such as wholesale, retail, and utilities—exhibit similar patterns of a decline in the labor share, they found it difficult to emphasize the importance of trade as a factor for the decline in the labor share.
As evidence contrary to the third factor, they cited the common experience of a decline in the labor shares across countries despite differences in the levels and evolution of unionization and other labor market institutions.
Meanwhile, as empirical evidence to support their "superstar firm hypothesis," Autor et al. pointed to the following four observations: (i) there has been a rise in sales concentrations within four-digit industries across the vast bulk of the U.S. private sector; (ii) industries with larger increases in product market concentration have experienced larger declines in the labor share; (iii) the fall in the labor share at the industry level is attributable mainly to a rise in the market shares of specific firms rather than a decrease in the labor share of incumbent firms; and (iv) the largest fall in the labor share is observed in the industries with the largest increase in concentration.
Some may argue that the superstar firm hypothesis can only explain the case of the United States. However, Autor et al. showed that declines in the labor share is observable in similar industries in European countries, with 12 out of the 14 countries examined exhibiting a negative correlation between changes in market concentration and changes in the labor share, i.e., the greater the level of concentration in the industry, the sharper is the decline in the labor share.
Furthermore, they also showed that in six European countries for which firm-level data are available, overall declines in the labor share are more attributable to a rise in the market share of low labor share firms than to declines in the labor share at individual firms, as is the case in the United States. They have thus proven that the superstar firm hypothesis holds in Europe. University of Pittsburgh Professor Daniel Berkowitz et al. reported that the superstar firm hypothesis can also explain a decline in the labor share observed in China.
♦ ♦ ♦
Now, why has market concentration increased across a broad range of industries? Autor et al. confirmed that technological dynamism—rather than anticompetitive forces—is an important driver for this trend, noting that concentration is more pronounced in more dynamic and innovative industries characterized by greater patent intensity and higher growth in total factor productivity (TFP) (reflective of technological advancement).
Using data from 24 member countries of the Organisation for Economic Co-operation and Development (OECD) for 2000 onward, a group of OECD researchers led by Dan Andrews showed that productivity differences have widened between the top 5% of firms and the rest. They attribute the phenomenon to a slowdown in technological diffusion between the frontier firms and the laggards, noting that leading firms can better protect their advantages than they used to, creating a "winner takes most" situation.
Taking the speed of patent citations as a proxy variable for technology diffusion, Autor et al. also showed that industries with slower technology diffusion exhibit higher market concentration and a sharper decline in the labor share.
Why has the labor share declined in superstar firms? Further research must take place to answer this question. However, as Harvard University Professor Lawrence F. Katz points out, U.S. firms are reducing their payrolls and increasing outsourcing to contracting firms, temporary help agencies, independent contractors, and freelancers. It is not hard to imagine that such "fissuring of the workplace" operates to lower the labor share by enabling large firms to save on the wage premiums and reducing employees' bargaining power.
The impact of increased dominance by superstar firms on the entire economy is affecting some of the most unlikely areas. An analysis by New York University Professor Thomas Philippon et al. found that underinvestment in the U.S. business sectors since the early 2000s has occurred under the influence of the increased market concentration, noting that a decrease in investment is more attributable to leading firms rather than to those left behind.
The rise of superstar firms is a universal economic phenomenon occurring across the world. It is important to study its impact on the entire economy and possible government interventions as a new policy issue.
* Translated by RIETI.
September 14, 2017 Nihon Keizai Shimbun
October 4, 2017
Article(s) by this author
October 4, 2017［Newspapers & Magazines］
May 31, 2017［Newspapers & Magazines］
March 27, 2017［Newspapers & Magazines］
October 5, 2016［Newspapers & Magazines］
July 4, 2016［Newspapers & Magazines］