In a few pieces out this morning, I make some noise about how unresponsive wage growth has been to the tightening job market, using the wage data from the Establishment Survey, which covers the private sector workforce. This is the key figure, showing that while wage growth clearly accelerated from 2 to 2.5% as the job market tightened over the past few years, it has since stalled out.
But do other wage and compensation series agree?
A lot of people like to cite the Atlanta Fed Wage Tracker. This series has some clear advantages, but I’m not quite sure what it’s telling me, as I’ll explain in a moment. At any rate, it too kind of shows some recent flattening, having sped up from about 2 to 3%, where it’s been wiggling around since about 2015.
Most wage series compare snapshots across different time periods, but the Atlanta one measures the wage growth of the same workers over 12 months (really, workers employed in both time periods, e.g., this August and last August). Because of this, it will typically show more growth than other series, since it’s including an “experience premium,” the bump to wages some workers enjoy as they age (or, similarly, as they add another year of tenure at their job). Also, relative to the snapshots, it’s less moved-around by compositional or demographic changes.
The unusual aspect of this wage series is that the median plotted above is not the median wage. It’s the median growth rate. They calculate the wage growth of people in the survey over the course of a year, and construct a distribution of growth rates from which they plot (a 3-mo. moving average of) the median. That means the median growth rate in any given month could be for low-, high-, or mid-wage workers. I find that a little confusing, as it doesn’t link to the earnings of what we typically think of as the median worker.
The next figure shows the yr/yr growth of the Employment Cost Index for compensation and wages. The ECI holds the industry and occupation mix constant and thus increases should not reflect changes in the shifting mix of jobs. It’s jumpy towards the end of the series, and there’s maybe a lift from around 2 to 2.5% growth. But, again, not much by way of acceleration.
Finally, there’s our own mash-up series, the result of principal component analysis of five different series: compensation from the productivity and ECI data, ECI wages, median weekly earnings, and the production-worker wage from the Establishment Survey. It’s shown here with a 3-quarter rolling average to smooth out its bips and bops. It shows a slow acceleration from 1.5 to 2.5% (yr/yr), but like the others, probably not as much as you’d expect from an unemployment rate that’s come down from 10% at its peak to 4.4% last month.
All the series show some responsiveness to the tighter job market, which is, of course, a positive development. That said, I don’t think the lack-of-wage-responsiveness story can be dismissed by referencing other wage and comp series.