The jobs report for March came out Friday, and the top-line number — the one that puts everyone who follows this on shpilkes — came in low. Analysts expected about 180,000 added payroll jobs, but the report offered 103,000. But, sticking with the Yiddishisms, not to worry! At least not yet, by a long shot. These monthly data are noisy — last month, payrolls were up 326,000 — so you must smooth out the noise and boost the signal by averaging over a few months. In fact, the pace of job gains in the first quarter of this year was a very solid 200,000, and the unemployment rate, at 4.1 percent, remains at a 17-year low.
But what about wage growth? You hear about each jig and jag in the stock market every five minutes, but the vast majority of working-age families depend on their paychecks, not their stock holdings (more than 80 percent of the value of the stock market is held by the richest 10 percent of households).
In sum, the tight labor market is certainly helping to boost wage growth, especially in places where it’s super tight. But, at least by the wage data out Friday, wage growth is slower than has typically been the case at such low unemployment. Over the past year, average hourly pay, before accounting for inflation, was up 2.7 percent. For middle-wage workers (see data note below), however, the increase was 2.4 percent (white-collar pay was up 3.2 percent). With inflation running around 2 percent, them’s some pretty measly real-wage gains for mid-wage workers.
The figure below underscores the point I just made about expecting faster wage growth at such low unemployment. It plots yearly, nominal wage growth for middle-wage workers in each month since 1964, when this wage series becomes available, that the unemployment rate was between 3.5 and 4.5 percent. Overall, wage growth in these months averages about 4 percent, but there are really three different clusters, or wage-growth regimes, in the figure.
Most recently, we’re in the bottom cluster, circled in the figure, with wage growth between 2 and 2.5 percent. The larger clump of dots in the middle clusters around the mean of 4 percent, and the top cluster hovers around 6 percent.
Why is our regime such an outlier to the low side? The next figure gets at that important question.
When comparing wage growth across long periods of time, it’s essential to grapple with key differences in the factors behind the numbers. The demographics of the workforce, inflation and productivity growth all influence wage growth, along with the most important factor: demand for workers and their bargaining clout. Regarding that latter point: One reason we expect and see stronger wage growth at lower unemployment is that tighter job markets force employers to bid wages up to get and keep the workers they need to meet strong consumer demand.
But the point of the above scatterplot is to kind of control for the key determinant of demand by holding the unemployment rate between 3.5 and 4.5 percent, i.e., low. So, let’s look at what else is going on.
The bar chart tells the story of the three different regimes from the perspective of inflation and productivity. For the middle group of bars, corresponding to the middle clump of dots in the scatterplot, fast productivity growth of 3 percent per year helped fuel nominal wage gains that trucked along at almost 4 percent.
For the top cluster, when wages grew almost 6 percent, inflation was also highly elevated, as high prices and high wages fed back into each other. These were the conditions in the late 1970s, when Paul Volcker shut down wage and price inflation with large, recession-inducing interest rate increases.
The bottom cluster — the one we’re living through — reflects both low inflation and low productivity. In other words, yes, we’ve got fairly strong labor demand, but low inflation and especially low productivity are putting downward pressure on wage growth. (One wrinkle here is that the unemployment rate is not quite comparable to earlier rates of similarly low magnitudes because more people have left the labor market and aren’t counted in the jobless rate.)
But does that mean workers, especially these mid-wage earners, must resign themselves to wage stagnation relative to these other regimes? Anyone who reads this column knows my answer: of course not!
True, economists don’t know how to pump up productivity, but that’s not the only path to faster wage growth. There’s also redistribution from the inflated profits share of national income to the labor share. By maintaining and even allowing further tightening of the job market (I’m talking to you, Federal Reserve), we can create the pressure for faster wage gains that shift income from profits to wages.
I know, stock market investors will hate this even more than trade wars. But it is precisely the rebalancing that we should expect to occur in a truly full employment labor market. In fact, the absence of such an economically equalizing rebalancing is one signal that we’re not yet at full employment.
By middle-wage workers, I’m referring to the 80 percent of the workforce that’s blue-collar in factory work and nonmanagerial in services. The analysis in the second figure uses quarterly data, as productivity is available only on that basis. While we now have unemployment and wages for the first quarter of this year, I had to forecast inflation and productivity for 2018 Q1 (though for inflation, I have data on the first two months of the quarter, so I had to forecast only the March data point).