The September minutes continued the discussion related to the ongoing Review that began in earnest at the July FOMC meeting. The September discussion mirrored for the most part the discussion in July, but it included more detail on options for revisions to strategy. What we learned was incremental and did not indicate clearly where the consensus is going. Again there was a focus on makeup strategies, with more discussion at this meeting of considerations that weigh against such strategies.
The staff presented the FOMC with yet another set of simulations of rule-based average inflation targeting (AIT) and price-level targeting (PLT). Those strategies clearly fail the “evolutionary” requirement. Enough already! There were reservations that the simulations were “tied too rigidly” to the details of particular rules and assumed that private sector agents understood and found the new strategy perfectly credible—in effect, rational expectations. It seems that such simulations have, at a minimum, diminishing returns. But, more importantly, such simulations just distract from a discussion about realistic options.
While I believe makeup policies like AIT and PLT are not under serious consideration, “most” participants were “open to the possibility that the dual-mandate objectives of maximum employment and stable prices could be best served by strategies that deliver inflation rates that over time are, on average, equal to the Committee’s longer-run objective of 2 percent.” Such strategies “may require aiming for inflation somewhat above 2 percent when the policy rate was away from the ELB [effective lower bound], recognizing that inflation would tend to be lower than 2 percent when the policy rate was constrained by the ELB.” We have described such strategies as being in the spirit of strict makeup strategies like PLT and AIT, but more flexible, allowing more discretion.
While most participants supported strategies that deliver an average inflation rate consistent with the inflation objective, many had reservations about the staff simulations that were tied tightly to the details of specific rules and, in any case, showed only modest benefits from makeup policies. Perhaps the most important reservation was a fundamental one that applies to more than just the staff simulations: “The responsiveness of inflation to resource slack had diminished, making it more difficult to provide sufficient accommodation to push inflation back to the Committee’s objective in a timely manner.” That problem, the flatter Phillips curve, limits the success of current strategies, rigid makeup strategies, and all the possible more flexible strategies that participants showed more support for.
Participants also discussed a number of specific challenges that apply to makeup strategies, rigid or flexible:
▪ The effectiveness of such policies depends on the credibility of future policymakers following through on an earlier commitment, which is especially tenuous when those future policymakers are faced with inflation above the objective.
▪ Keeping rates too low for a long time could lead to excessive risk-taking. So true! While it was suggested that macroprudential policies could deal with this, I have argued that this is an illusion. This is a real challenge for which the Committee has no answer.
▪ The credibility of many options was lower than for threshold-based guidance.
▪ A fully symmetric makeup policy would unduly limit the policy response when inflation was above 2% amid signs of an impending downturn.
Participants suggested several alternatives to explicit makeup-based strategies.
1. Most importantly, using threshold-based guidance sooner and more aggressively—an example of an evolutionary revision, and one we have seen as likely to be adopted.
2. Responding more aggressively to below-target inflation than to above-target inflation. This is an example of a strategy that is asymmetric—a property we expect any revised strategy will have.
3. Several suggested adopting a target range to communicate that there will be situations in which moderate departures of inflation from 2% would be appropriate. Any range-based proposal should make clear that the midpoint of the range is the objective and that inflation settling at a point at either end of the range is inconsistent with achieving price stability. Unfortunately, that caveat is not clear in some of the range proposals that have been suggested.
4. A couple suggested setting a range where the midpoint is modestly higher than 2 percent. This seems to be a strategy that sets an inflation objective that is moderately higher, inconsistent with their decision to remove such a revision from consideration in the Review. Even doing so over the short to medium term would seem to conflict with that decision.
5. A couple suggested setting 2% inflation as a floor—apparently instead of as a ceiling, which is how some participants have said the FOMC seemed to be treating the 2% objective! That also seems like a strategy that calls for an average inflation rate above 2%.
Participants also suggested that revisions to the language in the statement on strategy, reflecting any revisions adopted in the Review, could address:
▪ The conduct of monetary policy in the presence of the ELB constraint.
▪ The role of inflation expectations in the pursuit of the inflation goal.
▪ The best means of conveying the Committee’s balanced approach.
▪ The symmetry of the inflation objective.
▪ The time frame over which Committee would achieve its inflation objective.
Such revisions seem too extensive to be incorporated into a statement that is designed to be clear, simple, and brief. A more likely outcome is a press release that lays out the revisions, with more modest revisions to the language in the statement (for example, adding references to the ELB and inflation expectations and redefining symmetry). Good luck with better conveying the meaning of the balanced approach! Specifying a time frame for achieving the inflation objective is a non-starter because it would depend on the magnitude of the departure of inflation from the inflation objective as well as the gap between the unemployment rate and the estimated NAIRU.
The discussion was mostly consistent with our view of the “end game” of the Review, without any clear consensus on the specifics.
▪ No rigid, rules-based makeup strategies—specifically no AIT or PLT.
▪ A makeup spirit, without a makeup “commitment”; rather, a more flexible approach that allows for more discretion.
▪ Earlier and more aggressive use of threshold-based forward guidance. This seemed to be a near-consensus view. I expect the threshold to be in terms of an inflation rate (for example, the objective) rather than the unemployment rate (anyone knows the NAIRU for sure?).
▪ A definition of symmetry that is more policy-relevant, that is, consistent with a flexible strategy with a makeup spirit.
▪ An asymmetric policy, one that is more aggressive when inflation is below rather than above the objective. Symmetric and asymmetric at the same time. OK, a communication challenge!