from the Cleveland Fed

— this post authored by Roberto Pinheiro and Meifeng Yang

In most developed countries, the share of output accruing to labor has declined over the last 20 years. However, the underlying reasons for the decrease may have differed in the United States and other developed countries. In this Commentary, we examine some of the explanations economists have proposed for the decline in the labor share and discuss how well these explanations account for the decline across developed countries.

In most developed countries, the labor share – the percentage of economic output that accrues to workers as labor compensation – has declined over the last 20 years. Economists and policymakers study changes in the labor share over time because those changes indicate how much of the growth in labor productivity is captured by workers through real compensation growth. Examining ways in which the labor share has changed in other developed countries may suggest the possible future path of the labor share in the United States. As Blanchard (1997) pointed out, developed countries tend to be similar and close to each other in terms of technological innovation, so differences in the evolution of the labor share should not persist over time.[1] In addition, studying changes in the labor share across countries may allow us to evaluate the generalizability of proposed explanations for the underlying causes of the changes.

In this Commentary, we compare changes in the labor share in the United States with changes in 34 other developed countries, including European Union (EU) member states, EU candidate countries,[2] and other countries belonging to the Organization for Economic Cooperation and Development (OECD).[3] We investigate two potential explanations for changes in the labor share over time across different countries. The first explanation suggests that changes in the labor share across time can be partly accounted for by structural changes in the sectoral composition of the economy and partly by movements in the labor share within sectors (see Elsby et al. 2013 for the US case). The second explanation suggests that changes in the relative price of inputs, particularly a decrease in the relative price of investment goods, may have induced firms to replace labor with capital, such as machines and software (see Karabarbounis and Neiman 2013). Other reasons for changes in the labor share have been suggested, such as deunionization, increases in globalization that induced more international trade and offshoring, and changes in demographics, among others, so our analysis is not exhaustive and does not rule out the effects of other variables. However, the explanations we focus on are those that appear in the papers most often cited in the literature.

Structural Changes in Sectoral Composition as an Explanation

According to Elsby et al. (2013), the decline in the US labor share happened because of two changes in the US economy. The first was a shift in the country’s sectoral composition from a high-labor-share sector (manufacturing) to a low-labor-share sector (services). Second, the labor share within the manufacturing sector itself declined over time, and this decline was not compensated for by an increase in the labor share within the services sector. Elsby et al. suggest that these combined changes led to the observed decline in the overall US labor share.

Figure 1 shows that the decline in the labor share was steeper in other developed countries than in the United States, and it started earlier. In fact, most research shows that most developed European countries started experiencing a decline in the labor share in the mid-1970s (see Commission 2007), while the decline began in the late 1980s in the United States (see Armenter 2015 and Elsby et al. 2013).