Trivial points on the labor share

Noah Smith has a Bloomberg View article going over possible reasons for the recent decline of the labor income share of GDP.   He assesses four hypotheses,  which he calls: China, monopoly, robots and landlords.

His piece is brief and maybe familiar to many of you, so I will not bother summarizing it, except to say that the thinks the first three forces may actually be parts of one big thing, while the last is perhaps separate.

I don’t have much to add to the basic argument he presents, but would offer just three trivial observations.

First, I was struck by his claim that, “For decades, macroeconomic models assumed that labor and capital took home roughly constant portions of output — labor got just a bit less than two-thirds of the pie, capital slightly more than one-third. Nowadays it’s more like 60-40.”

I think that is right.  The models made the assumption, although not necessarily for any particularly good reason, aside from it seeming temporarily to be supported by the data and perhaps simplified the math in some cases.

Second, I am not sure we all agree on what the data are in this case.  Perhaps it is the labor share of net value added, rather than gross, that “should” be stable.  Take a look at the picture below which shows two measures of the labor share, one with gross value in the denominator, and the other with net.  Note that the gross share is close to Smith’s 60%, which may be a happy fluke.

Screen Shot 2017-04-24 at 6.19.12 PM

I focus on the domestic operations of the nonfinancial corporate sector, rather than overall economy as measured in GDP, simply because these data a readily available, presented in a way that adds up, and relatively free of some of the abstractions found in the GDP.  But full disclosure: I think the issue I am about to highlight is most pronounced in the sector I choose to picture, and the data there are not free of controversy.

Anyhow, for the domestic operations of NFCs, the labor share of net is less than 1/3 as far off full-sample average as the share of gross.  This does not mean that the issue Smith is trying to assess is mostly a statistical illusion. I am not qualified to weigh in on that. I just find it interesting.  It probably has something to do with the average life of the capital stock moving around, which I think is something you would want to consider, although not necessarily by just netting out its effect.

Third, from an income equality perspective, I don’t think the main issue here is that something called “labor” is falling behind capital and rent, although that may be happening. Some owners of labor, capital and land, i.e. people, are doing better than others.  This is another area where I would not get too attracted to essences.