It’s easy to see the attraction of make-up strategies. First, they allow for greater stimulus at the effective lower bound (ELB) because policymakers are committed to raising the inflation rate by more than otherwise. As a result, they have to maintain the fund’s rate lower for longer. Second, they promote an average inflation rate at the FOMC’s 2% inflation objective, which is arguably a better definition of price stability than only an objective of 2% itself. As a result, such policies contribute to keeping inflation expectations anchored at the inflation objective.
But the effectiveness of implementing a make-up strategy depends on the slope of the Phillips curve. Specifically, the slope has to be steep enough to allow the Committee to carry out that strategy. If the slope is relatively steep, the Fed will be better able to achieve the objective, which makes it easier to keep inflation expectations anchored at the objective and enhances the credibility that the FOMC can also overshoot the objective.
A relatively steep Phillips curve is especially valuable when at the ELB. When inflation is low and the fund’s rate is constrained by the ELB, a central bank must be able to raise inflation fast enough to keep inflation expectations anchored at the objective.
The Japanese Example: With A Flat Phillips Curve and Adaptive Expectations, Make-up Policy Is Not Credible To appreciate the greater, and perhaps insurmountable, challenge when the Phillips curve is very flat, consider Japan: The unemployment rate is 2.5%, the Phillips curve is very flat, inflation has been persistently near zero, and inflation expectations have become adaptive and have collapsed toward the persistently low inflation rate. (A similar dynamic appears to be in play in the euro area.) To enhance the effectiveness of its forward guidance, the BOJ introduced a make-up element to its strategy. But makeup policies are likely worthless in this case. The prospect that the BOJ could achieve an average inflation rate of 2% over any reasonable period is minimal. The prospect that the unemployment rate could be reduced enough further to raise inflation to 2% is questionable. And so is the prospect that such policies could raise inflation expectations, which in turn would raise inflation to 2%. The BOJ simply appears out of business.
It is fashionable to say: let fiscal policy take over. Fiscal policy is indeed becoming more stimulative in Japan. But the fiscal policy does what monetary policy can do: increase aggregate demand and lower the unemployment rate. But it faces the same roadblock in raising inflation: a flat Phillips curve and adaptive expectations.
The Savior: A Nonlinear Phillips Curve
However, if the Phillips curve is nonlinear (with the slope becoming steeper as the unemployment rate falls– at least below some threshold) monetary (and fiscal) policy could, at some point, become more effective in raising inflation. It is hard to test this proposition in the U.S., given the lack of experience with very low unemployment gaps. Sorry to have to bring the NAIRU into the story here, but it is the unemployment gap, (rather than the unemployment rate itself) that is in principle linked to inflation. For example, that gap in the U.S. today–based on FOMC participants’ median estimate of the unemployment rate in the longer run–is only ½ percentage point, reflecting the decline in the estimated NAIRU as inflation has remained low even as the unemployment rate has declined. So maybe we have not tested the nonlinearity in the U.S. yet. Maybe.
Makeup Policy and the Review
The Framework Review is principally motivated by the search for innovations in strategy that will make monetary policy more effective at the ELB. Makeup policy is the most promising such innovation. But, as I have argued above, the promise of makeup policy depends on the slope of the Phillips curve and the flat Phillips curve suggests that makeup policies may not be, well, promising. That would be compounded to the extent inflation expectations are beginning to erode here.
Perhaps an analogy will reinforce this point. Let’s assume–as some have suggested–that the FOMC raises its inflation objective from 2% to 4%. The question here is how would the FOMC orchestrate a move from 2% to 4%. How much would the unemployment rate have to fall to raise inflation by two percentage points? Think 10 or more percentage points? OK, that is an exaggeration, maybe. Perhaps if a 4% inflation rate were instantaneously credible without any action by the FOMC, inflation would instantaneously move to 4%. Not likely. Inflation in empirical Phillips curves generally depends on some weighted average of past and forward
looking expectations. And given the limited ability of the FOMC to raise inflation by stimulating aggregate demand, inflation expectations would more likely to be adaptive rather than “rational”!
Let’s add the ELB. Make-up policies should enhance the ability of the FOMC to provide stimulus at the ELB, and hence, depending on the slope of the Phillips curve, to raise inflation. Again, the flat Phillips curve is the fly in the ointment. Makeup policies are seen as contributing to anchor inflation expectations at 2%. Again, depending on the slope of the Phillips curve.
Makeup Policy with A Flat Phillips Curve
Given that the Phillips curve is very flat, it would be very difficult to even return to the 2% objective, when starting from well below 2%. It would be even harder to overshoot 2%, even with makeup policies. Given that challenge, a makeup strategy might have little credibility. Indeed, the challenge of returning to even 2% could make inflation expectations erode and fall toward a low actual inflation rate.
Asymmetry with Makeup Policies
The slope of the Phillips curve also has implications for the challenge of offsetting persistent undershoots compared with offsetting persistent overshoots. Offsetting undershoots requires a decline in the unemployment rate, indeed to below the NAIRU to raise the average inflation rate back to 2%. The decline in the unemployment rate is consistent with an improvement in social welfare and the return to an average inflation rate at 2% might be more consistent with the FOMC’s interpretation of its price stability mandate in the revised framework.
On the other hand, makeup policies would call for sufficiently restrictive policies to lower the inflation rate for some time to below 2%, if confronted with persistent overshoots. That would require a rise in the unemployment rate, to a level well above the NAIRU, and to a degree that depends on the slope of the Phillips curve. With a very flat Phillips curve, the required rise in the unemployment rate would only be feasible if there were a sizeable recession, something that would be unwelcome, to put it mildly. Thus, makeup policy, if implemented symmetrically, would not be acceptable. I have suggested a makeup policy when starting from undershoots and opportunistic disinflation when starting from above 2% inflation.
This asymmetry also has implications for makeup policies following overshoots that offset undershoots. To offset undershoots, inflation must for a time rise above 2%. How to get inflation back down to 2% when the offset is complete? A recession again, or a long period of below-trend growth and rising unemployment rate? (A pattern that, in any case, always ends in recession.)
Compared to What?
I have considered the downside to makeup policies here. But makeup policies, nevertheless, maybe the best hope for strengthening the effectiveness of monetary policy at the ELB. However, the failure of such a major innovation would undermine the reputation of the Fed as well as damage its credibility when announcing other innovations to achieve the same objective.
That’s why I have suggested a more incremental and evolutionary approach: More aggressive forward guidance with an inflation threshold, perhaps set a 2% objective. This would virtually guarantee an overshoot following persistent undershoots at the ELB, but not insist on a recession to return inflation to 2%. The
the downside of such a policy would be that it would not move as close to an average inflation rate at the 2% objective as makeup policies would.
The effectiveness of makeup policies depends on the slope of the Phillips curve. The flatter the slope, the less effective make-up policies will be in raising inflation and anchoring inflation expectations at the 2% objective. A flat Phillips curve compounds the downside of makeup policies, necessitating a recession to correct a persistent overshoot of the objective. While makeup policies do have some attractive features, the Committee will have to take into account the downside of such policies, especially with a flat Phillips curve.