Should the Fed Change Its Target?: Interview with Michael Woodford

When Michael Woodford speaks, central bankers listen. The Columbia University economist is widely regarded as one of the world's foremost experts on monetary policy.

In September, at an economic conference in Jackson Hole, Wyoming, he made waves with a paper explicitly advocating a significant change in the way central banks work: Instead of focusing primarily on inflation, he wrote, they should aim for a target level of economic output (expressed in nominal dollar terms). This would give them a framework to provide extra stimulus -- and to temporarily allow higher inflation -- at times like the present, when the economy needs to catch up to its previous trajectory in the wake of a deep recession.

Since then, the Federal Reserve has taken extraordinary steps to provide added stimulus. Most recently, it explicitly stated that it will keep interest rates near zero as long as the unemployment rate remains above 6.5 percent and inflation holds below 2.5 percent. We spoke with Woodford at the annual meeting of the American Economic Association to find out what he thinks about the Fed's latest efforts.

Q: What was the thought process that led you to support nominal GDP targeting?

A: Actually, it's an approach I've been advocating for at least a decade, though in my earlier writing about this I referred to a more technical variant of the proposal (what I called an "output-gap-adjusted price-level target") rather than to nominal GDP targeting. The idea is to have a target criterion with two qualities: It must focus on the level of a nominal variable rather than its growth rate, and it should involve some combination of prices and economic activity. I settled on nominal GDP -- the dollar value of all the goods and services produced in the economy -- because it's a measure that a central banker can talk about and be understood by the general public, and it avoids contentious issues such as the correct definition in practice of the "output gap."

I see it more as a refinement than as a move away from inflation targeting. In practice, targeting only means having a definite goal for inflation in the medium to long run, not an exclusive emphasis on inflation in deciding the appropriate nearer-term path for the economy. It is emphasized in Ben Bernanke's writing about inflation targeting, and the Fed's official statement a year ago that announced its inflation target, that one should also take into account the projected path for real economic activity when making short-run policy decisions. I think the Fed should be more explicit about how exactly it expects to choose among possible short-run paths, lest the inflation target itself cease to be meaningful; and I think that a commitment to a nominal GDP target path -- one that is chosen so as to yield the desired inflation rate in the medium to long run, while also explaining how to balance the inflation outlook with the outlook for real GDP growth in the short run -- would be a good way to do that.

Q: To what extent have the Fed's latest moves -- stating the levels of unemployment and inflation at which it would consider raising rates -- brought its policy closer to what you propose?

A: They have moved closer in one important respect: They are now talking about future policy in terms of economic outcomes, rather than a calendar date that will determine when interest rates should begin to rise. And it has some flavor of nominal GDP targeting because it involves both inflation and the real economy. The problem is that it's not clear what the new policy means for the longer term. Will the Fed always push interest rates down to zero when unemployment rises above 6.5 percent, or is this just a temporary thing? I presume it's the latter, but then the temporary policy runs the risk of raising doubts about how policy will be conducted later. A nominal GDP target would explain why they're doing what they're doing now, and also how they would shift to a less accommodative regime in the future, once the "catch up" to the target path has occurred.

Q: What new kinds of risks might a nominal GDP targeting regime introduce?

A: The biggest risk is miscommunication. People might think you've discarded the inflation target, rather than simply choosing an appropriate nearer-term target that is actually intended to guarantee the target inflation rate over the long run. That could lead them to expect higher inflation, a development that would make it harder to avoid higher actual inflation. The risk is particularly great now, because all the Fed's extraordinary measures have created a lot of uncertainty about what its current policy really is. Nonetheless, given the fact that the Fed has already departed from conventional policy, I think that a move to a nominal GDP target could actually provide much-needed clarity. It's not too late for it to have a positive effect.

(Mark Whitehouse is a member of the Bloomberg View editorial board. Follow him on Twitter.)