Why the Fed Needs 'Extended Patience'
John Williams, president of the Federal Reserve Bank of San Francisco, gave a good interview to the Financial Times this week. Williams is seen as a moderate on monetary policy, a centrist on the dove-hawk spectrum. He thinks the economy is now close to full employment and the time for raising interest rates is getting "closer and closer."
Sam Fleming pressed him for fresh information on how the Fed is approaching its big decision. Here's a point worth emphasizing, one that illuminates why I think Williams is on the wrong track: It's "decisions" not "decision."
Analysts are fixating on the timing of the first rate hike. When this comes, though, you can bet it will be small; if it happens mid-year, as most expect, it won't be much of a surprise either. In other words, it won't make much difference. What counts is how quickly rates go up after this first increase.
In thinking about that, the Fed is guided by a notion that former chairman Ben Bernanke tried, usually in vain, to impress on financial markets: There's a difference between easing back on the accelerator and applying the brakes. Williams made a similar point.
Bernanke was talking about quantitative easing and the Fed's balance sheet. An enlarged balance sheet keeps delivering monetary stimulus even after the Fed stops adding to it with new QE. Likewise with interest rates: Monetary stimulus won't end with the first rate increase. Unless rates rise abruptly, which the Fed hopes to avoid, they'll be lower than neutral for a good while yet, and as long as that's true they'll keep providing stimulus.
If the Fed wants interest rates on a neutral setting as soon as the economy is back at full employment -- and if it intends to get from here to neutral gradually -- it will have to start raising rates before full employment is reached. Also, monetary policy acts on the economy with a lag, compounding the problem. For both reasons, the Fed can't wait too long. It has to move before the economy is back at full employment and inflation has climbed back to its target rate of 2 percent. So the argument goes.
Larry Summers and the Bloomberg View editors -- with whom, it goes without saying, I agree -- say the Fed ought to wait until inflation shows the whites of its eyes. That formula was put to Williams and here's what he said:
Say middle of 2015 you started just removing accommodation, well the effects of that on inflation really wouldn’t be felt until mostly until the middle of 2016 or maybe 2017 given the lags in monetary policy. So when thinking about these decisions the "white of their eyes argument" is wait until inflation gets close to 2 per cent. The problem with that is you are going to significantly overshoot your target and then have to reverse course much more dramatically than if you were to start raising interest rates gradually in advance.
Williams has a point, of course, and a balance has to be struck, but the argument for extended patience, let's call it, is strong.
After a severe and prolonged recession, defining "full employment" -- the trigger for rising inflation -- is harder than usual. Employment is rising at a respectable pace, but there's still an unknown amount of slack in the labor market: Strong job growth last month occurred alongside an increase in the unemployment rate to 5.7 percent, because displaced workers rejoined the labor force and began looking for jobs.
Other pools of disguised unemployment -- people working part-time who'd rather be working full-time, for instance -- remain to be tapped. And for a while yet, wages (and hence inflation) may respond more sluggishly than usual to rising pressure in the jobs market. Expectations of low inflation seem well entrenched and will bear down on wage demands. And many employers will be slow to raise wages as the recovery proceeds because they were reluctant to cut them during the downturn -- the phenomenon of "pent-up wage-cuts."
In the early stages of the recovery, the fear was that the rise in long-term unemployment was destroying human capital and rendering people permanently unemployable -- in effect, tightening the labor market and raising the so-called natural rate of unemployment to 6 percent or more. But unemployment has fallen faster than expected and wages haven't taken off. So estimates of the natural rate of unemployment have dropped to 5 percent or so, and might fall further.
All this suggests that standard measures of output and activity need to be viewed with caution as precursors of inflation. The Fed's preferred measure of core inflation is running at 1.6 percent -- well below target. Wages and benefits, measured by the employment cost index, rose 2.2 percent in the year to December -- still much less than the 3.5 percent (target inflation plus growth in productivity) seen as normal for the U.S. economy.
Letting both those numbers move up before starting to raise rates -- and raising rates, once that process begins, slowly and gently -- does involve some risk of overshooting the inflation target, for the reasons Williams gives. The risk, on balance, seems justified.
It would be for the Fed to decide whether an overshoot, if it happened, required interest rates to rise "much more dramatically," as Williams put it. There needn't be any drama. The Fed has tolerated a prolonged modest undershoot of its inflation target. A slight overshoot should be no cause for alarm.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Clive Crook at email@example.com
To contact the editor on this story:
James Gibney at firstname.lastname@example.org