Free exchange | Monetary policy

No cushion needed, apparently

Fed officials prefer a dangerously low rate of inflation

By R.A. | LONDON

THE Federal Reserve basically never sees a recession coming (at least when it isn't busily creating one to whip inflation). In early 2008, when America was already in what would become the deepest of all postwar recessions, the Fed was still projecting quite healthy GDP growth in both 2008 and 2009. This blindspot is not as absurd as it sounds. A Fed that began anticipating a recession that it did not want to occur would essentially be advertising its incompetence (and, one hopes, generating some intense internal discussion over why more wasn't being done to prevent the foreseen downturn).

Nevertheless, recessions occur, with some frequency. At the moment the Fed is projecting quite healthy GDP growth through 2016. But that outlook could change, and it would take the longest postwar expansion on record for America to get through the end of the decade without another downturn.

It's not clear to me that Fed officials are planning ahead. In an interview with Financial Times journalist Robin Harding, San Francisco Fed president John Williams spoke as if America's most recent macroeconomic convulsion took place in the 1970s:

In his own economic forecast, Mr Williams said, the Fed will raise interest rates in the middle of next year with the unemployment rate at about 6 per cent, inflation at 1.5 per cent and “everything moving in the right direction”.

“At that point if we don’t start to adjust monetary policy there’d be a risk of overshooting,” he said. “You don’t wait until you’re at full employment before you start to raise interest rates from zero.”

That reads to me like the words of a man who has learned nothing at all from the experience of the past few years. That's probably a little unkind; what is said in public never corresponds exactly to what is said in discussions with the rest of the FOMC or in private. But I find this very troubling.

The Fed has spent the last 5 years and 3 months trying to figure out how to provide more demand stimulus to the economy when its main interest rate is near zero. It hasn't been entirely unsuccessful, but the process has been very messy. Monetary policy appears to have consistently underreacted to weak demand—delivering too little stimulus with too long a lag. That underreaction is down partly to a lack of familiarity with "unconventional" policy tools, and partly to FOMC members' concerns that unconventional tools involve risks that normal interest rate policy does not. I don't know exactly how much the zero lower bound has cost the American economy over the past half decade, but the bill probably runs to several trillion dollars.

So you're on the FOMC. You have plenty of recent, bitter experience with this important assymetry: nominal interest rates can rise as high as they need to in order to combat inflation, but they can only fall to zero before monetary policy becomes much less responsive to economic weakness. How do your views evolve? Not at all, it would seem, if you are Mr Williams, who appears to be suggesting that any risk of overshooting is intolerable. Better to put the current recovery at risk and court future disaster than treat the inflation target symetrically.

Or to put things somewhat differently: in his view, there is no point at which the value of the cushion against the ZLB provided by a rise in inflation over 2% exceeds the cost of that rise in inflation.

That seems extraordinary to me. I'm trying to imagine the worldview in which inflation at 2.5% or 3% is meaningfully more costly than inflation at 2%. We flatter ourselves in thinking that we can even measure inflation reliably enough to know that we've hit one of those targets rather than the other; in 2012 the margin of error for annual core CPI inflation was plus or minus 22 basis points, meaning that the 95% confidence interval covered a range of nearly half a percentage point.

But 50 or 100 more basis points of inflation do matter when the economy faces a demand shock. For a given real interest rate, an inflation rate that is 0.5% higher corresponds to a nominal interest rate that is 0.5% higher—and which can therefore be cut by 0.5 additional percentage points before the ZLB is reached. And once an economy is at the ZLB, an inflation rate that is 0.5% higher corresponds to a real interest rate that is 0.5% lower—more negative, and closer to the market-clearing rate. And not for nothing, a higher rate of inflation also translates into greater real wage flexibility and smaller employment declines for a given demand shock.

Why then would the Fed be so determined to raise interest rates next year, so as to prevent inflation from overshooting the target at all? It's worth noting that Mr Williams is not wildly out of line with the rest of the FOMC; futures markets are now pointing to a first rate increase, to 0.5%, in fall of 2015: only slightly later than Mr Williams would prefer.

The Fed should have learned more from recent experience. When nominal rates are at zero you do wait until full employment, or almost-full employment, before tightening policy. You do that because you want inflation to overshoot, because that is how you provide additional stimulus at the ZLB, and because that is how you avoid landing at the ZLB in the first place. There are many worse macro problems than inflation at 3%! Try overshooting for once!

Discover more

Religious competition was to blame for Europe’s witch hunts

Many children are still persecuted as alleged witches in Africa for similar reasons

Has BRICS lived up to expectations?

The bloc of big emerging economies is surprisingly good at keeping its promises


How to interpret a market plunge

Whether a sudden sharp decline in asset prices amounts to a meaningless blip or something more depends on mass psychology