by Timothy Taylor, Conversable Economist

It is a remarkable fact that the labor share of income in the United States hovered in the range of 60-65% of total income for 50 years–but has declined since 2000 and seems to be still falling.

Roc Armenter explores how this figure is measured and some possible explanations for the recent change in “A Bit of a Miracle No More: The Decline of the Labor Share,” published in the Business Review from the Philadelphia Federal Reserve (Third Quarter 2015, pp. 1-7).

Armenter reminds us that in the construction of income statistics:

“Every dollar of income earned by U.S. households can be classified as either labor earnings — wages and other forms of compensation — or capital earnings — interest or dividend payments and rent.”

Here’s the basic pattern of the falling share of labor income:

One possible explanation for this change involves alterations in how the statistics for labor and capital income are calculated over time. The single biggest factor here appears to be a change in how the income of self-employed workers is treated. The official statistics divide up their income as if the person was working a certain number of hours for pay, which is counted as “labor income,” while the rest of their income is “capital income” due to ownership of capital in the form of their business. But the way in which the Bureau of Labor Statistics does this division changed back in 2001. Armenter explains:

Indeed, until 2001, the BLS’s methodology assigned most of proprietor’s income to the labor share, a bit more than four-fifths of it. Since then, less than half of proprietor’s income has been classified as labor income. … [A]t least one-third and possibly closer to half of the drop in the headline labor share is due to how the BLS treats proprietor’s income.

Before thinking about why the labor share has fallen, it’s worth thinking about the remarkable fact that it didn’t change for such a long time. After all, the period from 1950-2000 sees a rise in the share of workers in service industry jobs, along with enormous growth in industries like health care and financial services. Surely, all of this should be expected to alter the labor share of income in one way or another?

Back in 1939, John Maynard Keynes wrote an article (“Relative Movements of Real Wages and Output,” Economic Journal, 49: 34–51) pointing out that the division between labor and capital income appeared to have been the same for the preceding 20 years in the data available to him. He points out five separate factors that should have been affecting the shares of labor and capital over time, and notes that apparently the changes in these factors are almost exactly offsetting each other, He characterized these closely offsetting effects as “a bit of a miracle“–a phrase that Armenter uses in the title of his article. The minor miracle of a roughly stable labor share of income for several decades after 1950 arises from its own array of offsetting changes in industries and in labor share of industries, Armenter explains (footnotes omitted):