This new spate of analysis, which I’ll describe in a moment, generates two important findings. First, considerable lingering labor market slack is still a drag on wage growth. Second, the linkages between wage growth and price inflation are not very tight at all. Both findings should lead those poised to snuff out wage growth — in the case of the Fed, by raising interest rates — to stand down.
A key challenge for the Fed in recent years has been figuring out just how tight or slack the job market is, a question that’s been harder than usual because the unemployment rate isn’t as revealing a signal as usual. The reasons for the weaker signal are weak labor force participation and unprecedented shares of long-term unemployment, both of which dampen the jobless rate’s traditional dominance as a measure of labor market tautness. Simply put, the job market is not as tight as the 6.2 percent unemployment rate implies.
But how does this relate to wages? A central theme in my own work is that one way you know that the job market is truly tight is when you see real wages consistently rising, especially for lower-wage workers. These are the workers with the least bargaining power, and just about the only time they’re going to get a boost is when employers must bid their pay up to get and keep the workers they need. Dean Baker and I wrote a book about this: In a job market with weak unions and high inequality, the working man and woman’s best friend is full employment. (Their other best friends are the Earned Income Credit — a wage subsidy for low-income workers — and a decent minimum wage.)
So, if you’re having trouble figuring out conditions in the job market, you need to evaluate wage trends, and as these recent papers reveal — and as you probably know if you’re drawing a paycheck — pay for most workers has been pretty stagnant for a while now. Yes, we go on about the bips and bops in stock prices every few minutes, but most working-age people depend on their paychecks, not their stock portfolios. In an economic recovery that’s more than five years old, isn’t it about time we saw some real wage gains?
Let’s look at the evidence. As you’d expect, there’s some improvement as the job market tightens, but not enough yet to produce real wage gains across the scale. I’ve tracked nominal wage growth on a quarterly basis using five different wage series and have shown that they’ve basically been stuck at 2 percent since 2009. Surely, a tightening labor market should show up somewhere in those series. And with inflation running at about the same pace, the implication is stagnant hourly wages.
As if to make my point, just yesterday the BLS released their monthly earnings analysis showing that over the past year (July 2013-July 2014), both average hourly pay and inflation rose 2 percent, such that the growth in real buying power was zero.
Second, two recent reports from Federal Reserve economists reveal the impact of slack on wage growth in interesting detail. In both cases, these analysts are assessing the argument made by Krueger et al that since a) short-term unemployment has returned to its pre-recession levels, and b) the long-term unemployed don’t add much to slack — they’re basically out of the job market but haven’t quite realized it yet — the job market is actually pretty tight.
Based on the significant and negative correlations between wage growth and labor slack, both studies reject that hypothesis, finding that as far as wage growth is concerned, the current job market isn’t that tight at all. The Chicago Fed study, for example, finds that short, medium and long-term unemployment all matter for real wage growth. They find the same thing for involuntary part-time work (i.e., part-timers who’d rather be working full-time), another significant source of slack. The fact that they fail to find this wage/slack correlation for voluntary part-time work further underscores their results.
Economist Jesse Rothstein recently drilled down on this real wage growth question in another useful way. Since it’s pretty uncommon for employers to reduce the nominal wages of their incumbent workforce, “wage rigidity may be masking trends in the wages offered to new hires.” So he looked at real wage trends in new hires across the full wage scale. From pre-recession years until now, he found that starting wages actually fell in real terms for about 85 percent of new jobs, with starting real wages down about 3 percent in about half the new jobs.
Finally, a new analysis from economists at the Cleveland Fed underscores a point the Dean and I have emphasized here at PostEverything: Even if wage growth does pick up, its connection to inflation is, as these authors put it, “more akin to a tangled web than a straight line.” They generate two important findings about that tangled web.
First, while slack and wage growth are still significantly and negatively correlated, that’s no longer the case between slack and price inflation. Second, it follows that wage growth doesn’t feed into inflation in a big way. To test this theory, the authors ask, “How much does is help to add wage growth to a model designed to predict inflation?” The answer: It doesn’t. The model’s forecasting accuracy is essentially unchanged by adding information on wage growth.
End of the day, all of this research yields bad news and potentially good news. The bad news is that five years into a recovery, wage growth remains far too weak and far too narrowly experienced. The good news, hopefully, is that the central bankers will take to heart the lessons from this new research: Though some economists are saying otherwise, considerable slack remains in the job market. Should wage growth accelerate, and this research convincingly suggests that it will as the job market tightens, that should not be assumed to be inflationary.
And if it helps, just put it to song: If wages start growing, it’s safe to “let ’em grow, let ’em grow … don’t slow ’em down anymore!”