It is a popular theme in news reporting that there has been a sharp decline in the labor share of income over the last four decades, and that this is a big part of the story of wage stagnation. The data don't quite agree.

Part of the confusion is that people often look at the labor share of GDP, a measure which includes depreciation of capital equipment. Since the depreciation share of GDP (the amount of spending needed to replace worn out or obsolete capital) has risen, that would definitionally lead to a fall in the labor share, even if there was no change in the split between labor and capital.

However, even looking at GDP, the loss of labor share would not explain much of the wage stagnation for typical workers over the last four decades. The labor share of GDP fell 2.6 percentage points over this period. This implies that wages would have been 4.8 percent higher in 2019 if the wage share had remained constant over this period. That's not trivial, it means someone earning $20 an hour today would instead be getting $20.96 in a constant shares world, but it is not most of the story of wage stagnation.

But if we want to be more accurate and pull out the impact of rising depreciation, the labor share has only fallen by 1.6 percentage points over this forty year period. That would imply wages would be 2.5 percent higher in a constant share world. That translates into a wage of $20.50 an hour for the worker now earning $20.00 an hour.

It is also worth noting that all of the fall in the labor share has occurred in this century. In fact, I would say that it is really a Great Recession story, where the loss in shares is overwhelmingly the result of the weak labor market in the years 2008-2012. In the last few years, the labor share has been rising as the labor market tightens.

It is true that the data show a drop in shares prior to the Great Recession, but it is important to remember that these were the housing bubble years. During this period banks and other financial institutions were booking enormous profits on loans that subsequently went bad, leading to hundreds of billions in losses in the years 2008-2010. In other words, much of the profits booked in these years were not real profits. If we corrected for the tidal wave of bad loans, it is not clear that there would be much left of the rise in profit shares in the years 2002-2007.

In any case, it is clear that the vast majority of the upward redistribution was within the wage distribution, with pay that used to go to ordinary workers instead going to CEOs and other top executives, Wall Street financial types, and highly paid professionals (e.g. doctors, dentists, and lawyers). If we want to reverse this upward redistribution, the first step is to be clear on who got the money.

This doesn't mean that we have not had failures in anti-trust policy, especially in areas like cell phones and Internet and cable service, but this is not the major cause of the upward redistribution of the last four decades.