Why Has the Fed Given Up on America's Unemployed?
by Adam S. Posen, Peterson Institute for International Economics
Op-ed in the Financial Times
August 20, 2013
© Financial Times
Complaints about public officials' short time horizons are well rehearsed: The gripe is usually that too many activist policies result from pandering to voters and special interests. But the reality is that measures are often discarded before they have a chance to work. Then, having not really tried, policymakers claim that the target was unattainable. In macroeconomics, it is the unemployed who suffer most from this repeated failure to follow through.
There is a rush in the United States and Europe to prematurely declare stimulus policies ineffective at reducing unemployment. Much of the persistent joblessness is deemed structural and the costs of addressing long-term unemployment too daunting. Labor regulations and skills mismatches clearly play some role in keeping the jobless out of work, but their impact is exaggerated to excuse inaction.
What a reversal from just a year ago. Last summer the US Federal Reserve's Jackson Hole conference ended with a paper by Edward Lazear, a prominent Stanford economist, who argued that almost all US unemployment was cyclical and thus reparable through stimulus. That conclusion was then welcomed by Ben Bernanke, the central bank's chairman. It was consistent with the Fed's insistence that the level of employment that can be sustained by the US economy without stoking inflation—the "NAIRU," the non-accelerating inflation rate of unemployment—had not risen. Policymakers had not yet accepted high unemployment as a fact of life.
That was justified—and was consistent with the rate-setting Federal Open Market Committee's (FOMC) subsequent decision to keep short-term rates near zero so long as unemployment was above 6.5 percent and expected medium-term inflation was below 2.5 percent. It injected stimulus, expecting that it would bring unemployment down.
Yet for the past three months, despite little improvement, most members of the FOMC have spoken about a desire to "taper" off stimulatory policy by September. They reaffirmed that desire even after the market interest rates rose almost 100 basis points as a result of their statements. This credit market tightening, combined with the ham-fisted fiscal sequestration—the across-the-board cuts in federal spending that took effect in March—have dragged on US growth.
Weak growth will not reduce unemployment. The FOMC members' own forecasts for falling joblessness—which are used to excuse inaction—reveal they assume that the more than 4 percent of US workers who dropped out of work after 2008 will not return to the labor market. That is unjustifiable defeatism.
Yes, there has been a demographic shift leading to falling growth in the female and immigrant labor forces—but these trends were identified before the crisis, and could explain at best a decline of a few tenths of a point. Some US workers received extended unemployment benefits, and others have moved on to disability rolls—but neither explain much of the higher unemployment rate. Most discouraged workers should still respond to a sufficient and sustained increase in aggregate demand.
So there is no reason to hold back on trying to drive US unemployment down through monetary and fiscal policy. An elastic supply of labor will keep wage growth low, which will suppress inflationary pressure. There is some evidence of mismatch between vacancies and jobseekers' skills at the aggregate level but Peter Diamond, awarded a Nobel Prize for work in this area, has warned against reading too much into this. There is little direct evidence of skills shortages.
While a precise breakdown of unemployment into cyclical versus structural components is impossible, policymakers need not worry. If anything, fixating on false precision of any labor market estimate is nothing but an excuse for inaction. Firstly, over or undershooting the NAIRU temporarily does not lead to explosive changes in inflation.
At present, there is no danger of a 1970s-style wage-price spiral. Back then we had different labor market institutions (such as unions and inflation-indexing of wages) and bargaining power (pre-globalization), as well as weaker central bank independence, which contributed to inflation. Marked growth in wage demands are unlikely. Those in jobs today are so eager to keep them that it is likely underemployment—working too few hours to make ends meet—will be a bigger problem.
Second, there is no way to find out the sustainable level of employment without experimentation. This was exemplified by the great triumph of the Alan Greenspan era at the Fed: letting the economy run hotter for longer in the mid-1990s than previous estimates of the NAIRU would have allowed and discovering that employment could rise without inflation. True, that experiment was based on Greenspan's insight that there had been a rise in underlying productivity growth—and its growth may now be slowing. But the relevant question for policymakers then, as now, was how much labor sits unemployed, and the ability of that labor to start working again.
The costs of pushing a bit too far are small and reversible. But the costs of letting unemployment persist are vast. Even reforms to reduce structural unemployment, which worked in Germany a decade ago or in the United States a decade before that, only take effect in an expanding economy. There is no good reason for the Fed to give up on the labor market—and thus no good argument for allowing the de facto tightening of monetary conditions to stand.