The first question is fallen as a share of what and from when? Some research has taken labor income as a share of GDP or profits as a share of GDP and that is simply the wrong denominator. To keep the capital stock intact, some output has to go to replacing depreciated capital so it is very important to subtract depreciation from the denominator and look at labor compensation as a share of national income or net national product. In the 1950s depreciation averaged 11.8% of GDP but over the last 10 years it has averaged 16.0%. We now buy shorter-lived assets that depreciate faster. The second question is fallen since when? The labor share of national income averaged 62.1% over the last four quarters. This is exactly the same share as the 15 years that ended 1965 Q4. Those 15 years saw an average CPI inflation rate of 1.5%,, an average unemployment rate of 4.9%, and an average real GDP growth rate of 4.0%. I doubt anyone could say that this was a bad economic outcome. Over the next 25 years the labor compensation share of US GDP averaged 65.5%. Inflation over these 25 years averaged 5.9%, the unemployment rate averaged 6.2% and real GDP growth was 3.1% (respectable but well below 4.0%). The 1950s and early 60s were, therefore, a relatively low labor share economy with booming growth and very subdued inflation. The current environment has a similar labor share and until recently low inflation but growth has been barely above 2% (until the last couple of quarters). The difference between the two periods is productivity growth and I am not sure that the labor share should be an independent variable in any economic model. The economy grinds out prices and wage rates and employment and production and the labor share of national income is a derived variable. It is a cyclical variable and as the labor share of income rises (the profit share falls), we are typically near the end of the expansion and the labor share typically peaks during the recession.
