The likely strategy in the next recession will be a swift decline to zero bound, perhaps forward guidance that is implemented more aggressively and more quickly, and is less conditional, closer to a more effective commitment strategy. Perhaps less reliance on asset purchases. No change in the inflation objective. Negative rates now on the table but down the list and in any case there are limits here as well. Between the proposals, we see a price-level targeting regime, while retaining a 2% (average) inflation objective, as the most attractive and less dramatic change in framework than nominal income targeting or a higher inflation objective. We are, in any case, in the midst of an extended period of research and more open discussion inside and outside the FOMC over the next year or two. And an overwhelming consensus among FOMC participants must be firmly established before a significant change in the monetary policy framework is made. As a result, we are not close to a change. There is no urgency now. We expect there might not be a change in the framework before the next time we end up at the zero bound. That might be a good time to implement a price-level targeting regime, and, as Bernanke has said, that framework could even be limited to such circumstances.
There is a vigorous debate about how Fed strategies should be adapted to a world of low real rates where the nominal fund’s rate will more likely be driven to the zero bound in the face of an adverse demand shock. Possible alternative frameworks include a higher inflation objective, price level, or nominal income targeting.
The point of departure is that in a typical recession the Fed cuts the nominal fund’s rate 5 percentage points. If the nominal neutral rate is only 3% (1% real), a five-percentage-point cut is not possible. As a result, we may be at the zero bound in almost all recessions going forward, with a resulting large output loss over time.
The second consideration is that there are questions about the effectiveness and potential of the unconventional policies implemented after the Great Recession. With asset purchases, less room today—both because it would start from a higher balance sheet and an already-low ten-year Treasury yield—effectiveness would be less than when the market’s not disrupted, with possibly diminishing returns. With forward guidance, there may little credibility that the FOMC would really accept material and extended periods of higher inflation, the prerequisite of effective forward guidance. But this is also a prerequisite for price level or nominal income targeting. So the task is how to make forward guidance more credible, likely by making it less conditional.
The clearest and most direct direction is to raise the inflation target to increase the buffer and provide more room for the FOMC to lower the fund’s rate. With a currently near-1% real neutral funds rate, it would take a 4% inflation objective to allow a five-percentage-point decline in the fund’s rate in response to a large adverse shock. But here we are concerned with what we should anticipate with new frameworks. We have long believed this is off the table, too big a step for a central bank to take except as a last resort. That’s because a 4% inflation objective could not possibly be viewed as “price stability,” not to mention it would be a nonstarter in terms of the public’s and Congress’s acceptance. And it might be regretted when r-star rises in the future. And, of course, if the FOMC is having trouble getting to 2%, how would it ever get inflation to 4%? Finally, it is more challenging for one country to do it alone. Expect a joint announcement at Basel if this is the direction taken. But not the next step, if ever.
Rosengren’s suggestion for a range for the inflation rate target rather than a point is interesting but in the context of a point inflation target and an expression of the range of tolerance, especially important following
large adverse shocks. This is, in effect, a nonlinear rule that provides a zone of tolerance that gives way to a more aggressive response when inflation moves outside the range.
Price level or nominal income targeting appear to be strategies that would better preserve the effectiveness of monetary policy in a low r-star world, reducing the prospect that we would get to the zero bound and increasing the amount of stimulus if we got to the zero bound. For price level targeting, it can be understood as a superior approach to promoting a 2% inflation objective than a point objective, as it is more consistent with achieving an average 2% inflation rate over some period of time than a point inflation target. It could work by stabilizing long stabilize long-term inflation expectations, which in turn provides a gravitational pull back to the inflation objective. But its main effect would be as part of a commitment to a lower-for-longer strategy, perhaps more effective than output-based or calendar-based forward guidance. Can this be communicated in a way the public (and the Congress) understands? Maybe. But, as Bernanke pointed out, what makes it effective if is if markets believe it. Importantly, the difference between price-level targeting and inflation targeting regimes might be very small in normal modest cycles but would become very useful in the case of a dramatic adverse shock.
In nominal income targeting, the FOMC targets a level of nominal income that rises at the sum of the estimated rate of potential growth and the FOMC’s inflation objective. Its advantage relative to price-level targeting is that it responds to both sides of the mandate; that is, to a slowdown in real growth as well as a decline in inflation. Once again, its main value comes from its signaling “lower for longer.” Perhaps less effective at stabilizing inflation expectations. More difficult to explain to the public. And not a target that can remain in place for long or even medium-term, as it will have to be adjusted in response to changes in potential growth, and lags in making such adjustments will result in policy errors. But, as Summers has argued, this may be a more palatable way of introducing a 5% inflation objective, as the target nominal income could grow to the sum of potential growth plus 5%. In this way, though sneaky, it might dominate a higher inflation objective itself.
But a price-level targeting framework keeps the inflation objective more explicit, might be more comfortable for central banks and might work well enough to allow the FOMC to maintain its 2% inflation objective. And a price level target would also simply replace the inflation target in a policy rule which would still prescribe a systematic response to deviations from full employment. And, in practice, for modest adverse shocks, there might be little actual difference from an inflation target.
While there seems to be less enthusiastic about negative rates in the U.S., that direction is attractive because it would simply extend the range for traditional monetary policy beyond the zero bound. Easy to do and easy to explain. Still, there are questions about costs relative to benefits, and public acceptance if passed through to depositors.
Finally, political feasibility is a hurdle for any alternative policy framework. Not only must Chair forge an agreement on the framework within the FOMC, he or she must also be willing and able to defend it before Congress and the U.S. public.