Wednesday, September 25, 2013

Further Notes on the Declining U.S. Labor Share

I've posted a couple of times recently about the fact that the labor share of total income has declined in recent years, both in the U.S. economy and around the world. But interest in the topic seems strong, and there's an intriguing new paper out on "The Decline of the U.S. Labor Share," by Michael Elsby, Bart Hobijn, and Ayşegül Șahin in the latest issue of the Brookings Papers on Economic Activity.  Here, I'll list some of the overall themes that come up both in this paper and in previous posts, together with new insights from this paper.

1) The fall in labor share of national income in the last couple decades is quite a remarkable change, because for most of the 20th century, the labor share of national income could be thought of as pretty much a constant. Back in 1939, John Maynard Keynes was writing that the “remarkable constancy” of the aggregate labor share is “a bit of a miracle.” Paul Samuelson made a similar comment in a 1949 article. Nicholas Kaldor documented a similar result in the late 1950s. When I was first learning economics in the late 1970s and early 1980s, the labor share of national income was typically viewed as more-or-less constant, with some variation over the business cycle Indeed, I spent a certain proportion of time in my college years pointing out that the facts at that time did not support the claim that capital was receiving an ever-greater share of national income.

2) The seeming consistency of the labor share of national income up until the mid-1980s was actually the result of trends that counterbalanced each other. The shift from manufacturing to services tended to push down the labor share over time, but the growth of the health care industry tended to push up the labor share over tiem.

3) The decline in labor share is not the same as the rise in inequality of incomes. All of labor income is included in labor share, including what is earned by the top 1% as compensation for labor.  Elsby, Hobijn, and Șahin write: "[T]he labor share in the United States has in fact been buoyed up increasingly by the rising income shares of very rich workers, like CEOs for example. Thus, the decline in the labor share conceals, rather than exposes, most of the large increases in inequality that have emerged in recent decades."

4) The decline in the labor share of national income is occurring, but is also probably overstated in the commonly used statistics.  Elsby, Hobijn, and Șahin point out that the usual statistical approach has no easy way of estimating the income of those who are self-employed, and as a result an assumption buried in the data is to assume that the hourly compensation of payroll employees and the self-employed has changed in the same way over time. However, they believe that the pay of the self-employed has probably increased in the last couple of decades relative to the pay of payroll employees, and so the labor share is a little greater than the official statistics show. They find that "around a third of the decline in the headline measure   is a symptom of the method used to impute self-employment income."

5) One potential explanation for the declining labor share is that an increased rate of capital investment substituted for labor. However, Elsby, Hobijn, and Șahin find that the evidence does not support this explanation; specifically, the amount of capital/output wasn't rising over this time.

6)  Another possible explanation for declining labor share is the decline in private-sector unionization rates in recent decades. owever, Elsby, Hobijn, and Șahin find that the evidence does not support this explanation; specifically, the fall in labor share across industries doesn't seem much correlated with the changes in unionization rates in those industries.

7) Yet another possible explanation for the declining labor share is increased global competition.  Elsby, Hobijn, and Șahin write: "Our analysis identifies offshoring of the labor-intensive part of the U.S supply chain as a leading potential explanation for the decline in the labor share." More specifically, they document a close connection between which industries are exposed to greater import competition and which industries saw the biggest fall in labor share during the last 25 years or so.