If the Fed Could Turn Back the Hands of Time

Perhaps that’s something The Maestro might have thought he could do, but the members of the Federal Reserve are showing considerably more humility these days. When you read their words, it seems they are by and large trying to do their work, and in a stretch when possibilities formerly known as curiosities, like the Zero Lower Bound, hopped out of theoretical research papers and onto their plates.

In a recent interview, David Beckworth asked St. Louis Federal Reserve Bank President James Bullard if, could they start all over again, members of the FOMC might consider nominal GDP or price level targeting instead of the current dual mandate of stable prices and hard-to-define full employment, Bullard’s careful response was, “I think…from just a theory basis, I think there is more sympathy than there would have been a few years ago,” and added we aren’t there yet. He himself is more sympathetic than he once was.

Inflation? Follow the Trend…

Bullard brought up the question–what drives inflation?—as a key issue today, noting that, “…the answer you get from various angles is always the same: that inflation doesn’t seem to be related to the variables that we think it should be related to.” Although he advances inflation expectations as one of the stronger contenders, even that isn’t particularly well correlated with inflation. (He also likes the idea that the central bank target moderates expectations, while the bank drives the ‘flation car.)

The Phillips Curve seems to have lost its swag, Bullard maintains it never had much, and the (unorthodox and troubled) fiscal theory of the price level—that prices are determined by government debt and fiscal policy, with monetary policy playing a muted second fiddle—“leads to a lot of confusion about what is being said.”

A funny thing about the Phillips Curve: it’s been roundly questioned since it was knee high to a grasshopper. Just six years after A.W. Phillips laid it out, Milton Friedman and fellow monetarists pointed out various issues. To Friedman, it had a basic defect, the inability to distinguish between nominal and real wages, and held only in the short-run and when inflation and expectations were steady. The economics of the coming decades further tarnished the curve’s reputation.

A bit simplistic but, of course, if the Phillips Curve were truly robust, there would be no such thing as stagflation; we’d be in a naturally correcting environment. Yet in the 1970s unemployment increased along with inflation, and in the 1990s unemployment and inflation fell together. The many other economic factors that affect pricing, including tight competition, supply shocks, a greater central bank presence, and changes in the make-up of the labor market, over-power the logical connection between low unemployment and inflation.

“Deflating inflation expectations,” a recent study by Stephen Cecchetti, Michael Feroli, Peter Hooper, Anil Kashyap and Kermit Schoenholtz, suggests we should pay less attention to expectations and slack and focus instead on trend. They argue that while inflation expectations do correlate with the trend, it’s a misleading relationship because as inflation moves, expectations move with it. They point out that if inflation is moving at 2%, most will expect it will continue to do so, and will be right. They suggest instead we compare the change in inflation expectations to the change in inflation. That doesn’t work so well: changes in inflation expectations “tell us nothing” about changes in inflation.

Since inflation is currently a bit below the Fed’s target, and long-term inflation closely follows the past few years’ average, they suggest the Fed will likely need to allow a short, modest overshoot to get us to 2%.

But Bullard also points out that in 2012, when headline and core inflation were right on target, he thought the Fed’s “work was done,” but then inflation began to slip, which he has noted was happening again in the spring. That overshoot may prove elusive.

…and get rid of Core PCE

Bullard suggests the core measure of PCE should be “thrown out.” First, he is not willing to go back to his constituents and say, “We stabilized prices, just not the ones you have to pay the most often.” Second, one of his colleagues looked into food prices and found that for some stretches they are the least volatile of all components. Bullard called it “kind of comical” that we’re throwing them out because they were so volatile in the 1970s. Third, core inflation “sunk a bit, maybe a couple of tenths on a y/y basis,” when energy prices fell in the 2014, and have since reversed course. That’s because there are core components that are heavily affected by movements in energy prices, so the effect is still in there.

Bullard favors the Dallas Fed’s trimmed mean measure, running at 1.5% in May, but since so many models were built on core prices, he doesn’t think there is much will to recalibrate them all. We’d say this is not a good reason not to work to improve such an important measure, but it may be an overwhelming one. Perhaps that argues for nominal GDP, which washes out any issues with how inflation is forecast and tracked.