When will central banks in the United States and Britain start raising interest rates? The question preoccupies financial markets, and much turns on the answer. What's happening in the two countries' labor markets — and what the Federal Reserve and the Bank of England think is happening — is the crux of the matter.
In this sluggish and hesitant recovery, the first question is harder than usual to answer. Unemployment is the simplest measure of slack in the labor market, and it's fallen a lot in both countries. At about 6 percent of the labor force in each case, it's less than one percentage point above the rate central banks see as the long-term equilibrium — the rate consistent with stable inflation.
The turning point for inflation, according to this measure, is almost upon us, and the time for central banks to act is very soon. In a recent interview, Dallas Fed President Richard Fisher said, in effect, that the moment has arrived.
The problem is, other measures of labor-market slack aren't in line. Above all, there's little sign yet of rising labor costs. In Britain, make that no sign: There, real wages are falling, as David Blanchflower and Stephen Machin emphasize in an article for VoxEU. In the U.S., wages are rising only slowly; allowing for growth in productivity, the labor market is exerting no inflationary pressure.
How come? The theme emerging in presentations at a recent symposium at the Peterson Institute for International Economics is that central banks shouldn't gauge the state of the labor market by looking at measured unemployment. Some of the people who've stopped looking for a job (and therefore aren't counted as unemployed) will go back to work as the economy picks up. In addition, many people who want a full-time job are having to work part-time.
In the U.S., these forms of disguised unemployment add maybe two percentage points to labor-market slack. (In addition, even though output is rising, labor productivity remains lower than you'd expect, implying that some companies may have retained more labor than they need at current levels of demand — a third kind of disguised unemployment.) Add this to the one percentage point suggested by the gap between measured and equilibrium unemployment, and you have plenty of room for labor demand to rise without causing inflation. The corresponding figures for Britain are similar.
The answer to the second question — how close to higher inflation is too close? — is equally important and argues the same way. Even if the total employment gap were as small as measured unemployment suggests, central banks should be slow to raise interest rates.
Why? Because the depth of the recent recession has created greater-than-usual uncertainty about the way the labor market will affect inflation. In these circumstances, the costs of excessive caution in stimulating demand probably outweigh the costs of going too far: On one side, you have the permanent costs of elevated (measured plus disguised) unemployment; on the other, the temporary nuisance of having to move against wage-and-price inflation once it's visible, as opposed to merely anticipated.
Central banks are right to worry about inflation. That's their job. Even so, it would be better for them to keep interest rates at zero for too long than raise them too soon.