Paul Krugman and others keep coming back with this graph from the EPI:But there's a lot of empirical problems with that graph, discussed by Scott Winship here. Basically, the graph--and the EPI study it came from--compares apples to oranges, and then tries to correlate the difference to things other than the type of fruit.
Matthew Yglesias took some of those criticisms to heart and made this graph:One issue that comes to mind here though is that labor supply did not remain constant over that period, and in particular the period where the output/compensation ratio is rising was when women were entering the labor force, while the latter period where the ratio fell represented an era in which both the average work week and the labor force participation rate were in decline.
Here's another graph, this time comparing output per hour of work to the average wage for production and non-supervisory workers:It's very hard to avoid the impression that in 1980 there was a seismic shift from a high-wage economy to a low-wage economy. Front-line workers used to earn more than half of what they produce, now it's less than 40 percent.
Obviously, not all workers are "production and non-supervisory" workers. The series for hourly earnings of all workers doesn't go back far enough for our purposes, but I suspect we'd find that this is a story about inequality--automation and computers have elevated the status of managers and decreased the value of front-line labor. But what's so interesting about this graph is the sharpness of the change right around 1980. There was a period of adjustment from 1980 to the mid 1990s, to be sure, but that's still a pretty sudden and sharp change.
/cdn0.vox-cdn.com/uploads/chorus_asset/file/3911068/fredgraph-2.png)
