Last Friday brought the good news that the jobless rate fell last month to a 10-year low of 4.4 percent. According to various authorities that proclaim such things — the Federal Reserve, the Congressional Budget Office — that’s below full employment, meaning it’s lower than the lowest rate they believe to be consistent with stable inflation (hold that last thought, because it will be important later).
They may be right. The labor market has unquestionably tightened up a great deal over the past few years, and that’s led to faster wage growth, one of the key benefits of full employment. True, the jobless rate is an incomplete indicator, but as I pointed out last week, numerous other indicators, including the more comprehensive underemployment rate, also are flashing a tightness in the job market.
By these measures, we’re at full employment, which is worth noting when you consider how that has been the exception in the United States in recent decades. The figure below compares the CBO full employment rate to the actual jobless rate for every quarter since 1949. It shows that we used to be at full employment almost 70 percent of the time. Since 1980, we’ve achieved that state less than 30 percent of the time. Given the added bargaining clout that full employment delivers to middle- and low-wage workers, it’s no accident that real wage stagnation and much higher economic inequality evolved in the latter period.
[Important warning to econ students: The clearly wrong assumption that we are almost always at full employment underlies most theories in economics, especially the 101 stuff. Do not accept this. Bug your profs about the above graphic; I hereby empower you to be that annoying student!]
The question is, while I’m fully confident that these metrics measure what they say they do, are they accurate gauges of full employment?
I’m afraid they are not, at least in the sense that it has become increasingly challenging for economists to identify just how much slack is left in the labor market. That’s because the signal we’ve always tapped for this — the correlation between labor market tightness and wage/price pressures — just ain’t what it used to be. Both wages and especially prices just don’t react as strongly as they once did to tighter labor markets.
The figure below shows the growth rates of (a) core inflation and (b) nominal wages of blue-collar factory workers and non-managers in services over numerous periods of falling unemployment (since those periods are of unequal length, the growth rates are annualized so they’re comparable). As you can see, since the 1980s, those growth rates have been falling. In the current episode of falling unemployment, inflation in particular has been quiescent.
This has not gone unnoticed. A few years ago, when unemployment was 10 percent, economists were asking “why isn’t the rate of inflation falling more?” Now we’re asking, or at least I am, if we’re really at full employment, where’s the price pressure?
I asked a bunch of my colleagues how they’d answer this question. Economists Dean Baker and Elise Gould point to a few labor market indicators — quit rates and employment rates for working-age people — that are still somewhat depressed. Jason Furman notes that noninflationary wage gains are being paid for by shrinking profit margins, something I’d chalk up as another benefit of tight labor markets. Marshall Steinbaum and Heather Boushey argue that at least so far, even with tight labor markets, workers’ bargaining power remains too weak to create much wage pressure. Carola Binder suggested that inflation expectations are “well-anchored,” meaning that because people expect the Federal Reserve to enforce its 2 percent inflation target, the rate of price growth is locked in around that target (as Binder points out, this explains both why inflation didn’t fall earlier and why it’s not rising that much now). Ben Spielberg wonders whether the increase in Internet shopping is enforcing more price competition that dampens producers’ pricing power.
[A big “thanks” here to my smart, responsive colleagues!]
Equally interesting are the implications of these developments. I can think of two big ones.
First, the low correlation between tightness and price growth puts the Federal Reserve in a bind. It simply can’t know, at least with much confidence, the jobless rate commensurate with full employment. Chair Janet Yellen, who thinks deeply about this stuff, says the Fed wants to be data-driven, but that’s a lot tougher when the data linkages are weak. Clearly, the recent rate hikes, though they’ve been slight, show that the Fed believes we’re at or near full employment and that it’d better get ahead of the curve. But for all it or anyone else knows, the full employment rate could be not 4.7 percent, but 2.7 percent. It’s awfully hard to end up where you want to be when your map is practically invisible.
Second, if, in the spirit of Steinbaum and Boushey’s view, many workers will find it tough to claim their fair share of growth even at very low unemployment, we’re going to need offsetting policies, such as higher minimum wages, a much expanded earned-income tax credit, and direct job creation in places where labor demand is still weak (as per Baker and Gould’s analysis).
The structure of the economy has clearly changed a great deal in recent decades, and mostly in ways that tilt against those who depend on paychecks versus stock portfolios. Policy needs to catch up.