Thread by @jnordvig

Should the Fed lie?

Some argue that showing 2023 rate liftoff at the March 17 FOMC would be counter-productive, and that it would be better to manipulate the dots lower to strengthen the forward guidance.

Is lying sound central bank communication?

On the one hand, the Fed is trying hard to be accommodative, and the new core element in the Fed's framework is not to be preemptive, and instead wait for realized (inflation) outcomes, before embarking on tightening. Hence, you want low dots.

On the other hand, some FOMC participants may indeed think that inflation will hit their objective (2%, and on a path to exceed it) within the forecast horizon. Hence, you can argue, within the framework, that it is logical then to show 2023 lift-off in that scenario.

For the record, 2023 lift-off is not an idea that has been entirely taboo among FOMC participants, even after the new framework was adopted in August 2020. The last dot plot from December already had 5 dots above zero.

To be clear, everybody at the
Fed has signed up to the new framework (and its lack of preemptive tightening). Hence, the outlier-early-lift-off dots should be seen as reflecting those FOMC participants with the higher inflation forecasts (>2%), not opposition to the framework.

There will be be significant upward revisions to the Fed's forecasts. Certainly to the growth growth (fully embedding two rounds of BIG fiscal stimulus). And some feedthrough to the inflation forecast is also likely, although how much is uncertain [see the Jan minutes]

There is even more uncertainty about feedthrough to the policy rate path. Each FOMC participant may have slightly different view of what "inflation moderately above 2 percent for some time" means. But as the PCE forecast moves up, so should the dots (all else equal).

Should the Fed (the individual participants in connection with the dot plot) suppress their actual views on the likely policy path in the future, to enhance the forward guidance now? As some observers now seem to recommend?

That would ultimately create a credibility problem in the future, if the Fed decides to hike when the economy realizes a certain outcome (but had lied about what it would do in that scenario). It would shock the market then. Hence, a little gain now, would have a big cost later.

To enhance the forward guidance the Fed can, via the dot plot, reveal how much inflation overshoot it will tolerate. If it wants to anchor short-rate expectations more, it can put down higher PCE forecasts, with unchanged dots.

But it should only do so, if that truly (and honestly) reflect how they think about the reaction function under the new framework. And for some FOMC members that may indeed be the case. But then they have to stick to it.

There is a large literature on central bank credibility. While 2021 is about being pragmatic, expecting the unexpected, and de-learning false wisdom from the past, not all wisdom from the past is false.

Credibility takes a long time to build, but only a short time to undermine. I think central banks should be careful about lying. Not for religious reasons, but because their long-term credibility is key to them achieving their goals efficiently. END.