Monday, December 29, 2014

Tinkering On the Margins

Paul Krugman disagrees with a point I made in my last post. Specifically, he takes issue with my claim that a monetary regime change is needed for both monetary policy and fiscal policy to effective at the zero lower bound (ZLB). To make my case, I gave as an illustration a scenario where the U.S. Treasury Department does a helicopter drop that is offset by the Fed tightening once the helicopter drop starts to raise inflation. I also said a helicopter drop in the Eurozone would face a similar fate from the ECB. Krugman thinks this is wrong: 
What Beckworth seems to be saying is that the Fed and the ECB are at their inflation target, and would therefore tighten policy if the economy were to expand and inflation to rise. But they aren’t at their inflation targets! The Fed has been below target for a number of quarters; the ECB is way below target. 
Krguman's argument, then, is that fiscal policy could raise aggregate demand up to the point where it raises inflation to its target. So why not have a helicopter drop? Surely it would do some good, says Krugman.

Okay, let say for the sake of argument Krugman is correct about the Fed not being able to hit its inflation target. If so, this still only amounts to fiscal policy tinkering on the margins. The Fed's preferred inflation measures, the PCE deflator and core PCE deflator, have both averaged about 1.4% since 2009. So we are talking about 60 basis points of wiggle room for fiscal policy to work. Do we really think that within this narrow window fiscal policy could have generated enough aggregate demand growth to close the output gap? 

To help us see that this is just tinkering on the margins, let us revisit the point Krugman and I agreed on in our previous posts: a monetary regime change is needed to make monetary policy effective at the ZLB. One example of such a monetary regime change would be a price level target that returns the PCE to its pre-crisis trend path. To return the PCE to its targeted path would require a temporary burst of higher-than-normal inflation. The expectation of and realization of this inflation burst would be the catalyst that spurred robust aggregate demand growth. 

Now let us pretend the Fed actually implemented a price level target back in 2010 when it began QE2. Specifically, imagine the Fed had made QE2 conditional on the PCE returning to its 2002-2008 trend path. The figure below shows this scenario with three different paths back to the price level target. Note that each path represents differing rates--5%, 4%, and 3%--of 'catch-up' inflation and for each path there is a significant amount of time--16 months, 26 months, and 49 plus months--involved to catch up to trend.

What this illustrates is that to get the kind of robust aggregate demand growth needed to close the output gap back in 2010, there needed to be a sustained (but ultimately temporary) period of higher-than-normal inflation. Doing more fiscal policy to squeeze out the last 60 basis points of the Fed's 2% inflation target would not cut it. Again, it would be tinkering on the margins. If fiscal policy really wanted to close a large output gap at the ZLB it too needs the support of monetary regime change.

Now in the Eurozone it is true that the ECB is further from its inflation target so that would give fiscal policy more wiggle room. But there is also a larger output gap in the Eurozone. So I think the tinkering on the margins critique applies there too. The Eurozone also need a monetary regime change to make fiscal policy really pack a punch.

But there is more. I think a reasonable case can be made that inflation is actually in the range where the Fed wants it to be. If so, then even the wiggle room is gone. Krugman anticipated this response from me:
The Fed has been below target for a number of quarters; the ECB is way below target. And don’t say that the failure to raise inflation rates shows that they must be happy with where they are. The whole point of our previous discussion has been that monetary policy is ineffective under zero-interest conditions unless you are willing to change regimes! 
Yes, in general, we need a monetary regime for monetary policy to be effective at the ZLB. However, there is compelling evidence the Fed has been doing QE to keep inflation in a range where it is comfortable. In fact, a number of observers have come to the conclusion that the Fed does not have a 2% inflation target but a 2% inflation ceiling. They note that a 2% target would be an average, and the Fed should be willing to allow inflation go above 2% as often as it is below. But that is not the case as noted below: 
[I]t turns out that the Fed’s 2 percent target for core inflation is not a target, it’s an upper bound. 
That’s not supposed to be how it works. If you really think that around 2 percent inflation is right... you’re supposed to view 1 percent inflation as being just as bad as 3 percent; in a situation in which inflation is below the target rate, you’re supposed to see a rise in that rate as a good thing. And correspondingly, if you’re where we are now, with below target core inflation and high unemployment, all lights should be flashing green for expansion. 
Instead, however, it’s clear that below-target inflation is considered no big deal, but that the Fed is extremely averse to seeing inflation rise above target, even temporarily.
That comes from none other than Paul Krugman. If this understanding is correct, then the Fed actually is targeting a range of inflation and is not undershooting its target. I previously presented evidence that this range falls between 1% and 2%. Let me briefly review it.

First, the timing of the Fed's QE programs suggests that the FOMC initiates them when core inflation is under 2% and has been falling for at least six months. It also indicates the FOMC tends to end QE programs when core inflation is above 1% and has been rising for at least six months. This can be seen in the figure below:

Second, the central tendency ranges of inflation forecasts provided by members of the FOMC consistently show 2% as an upper bound. Below are projections for 1-year and 2-years out:

It is remarkable that FOMC members are predicting inflation no higher than 2% two years out. Since the FOMC has meaningful influence on inflation this far out, this forecast reflects FOMC members' beliefs about current and expected Fed policy. They see the Fed doing just enough to keep core PCE inflation under 2%.

Fed Chair Janet Yellen admitted as much in her last post-FOMC press conference:
But it’s important to point out that the Committee is not anticipating an overshoot of its 2 percent inflation objective (p.13)
In short, inflation is below 2% in the United States because the Fed is happy with it being there. Fiscal policy is not going to change that and even if it could it would amount to tinkering on the margin.

A stronger case can be made that inflation is below the ECB's target. However, it is not clear how much it is below target since the ECB's definition of price stability is inflation under 2%. So maybe there is some wiggle room for fiscal policy, but nothing close to what is needed to close the output gap. 


  1. Excellent blogging. Yes, for the Fed the 2008 Great Recession has been an opportunity to wratchet down inflation, and the new target must be about 1.5% on the PCE.
    The price paid for a small decrease in a nominal price index was enormous in lost output aand employment.
    But the Fed is an independent agency, don't bother them with the consequences of their actions.

  2. Very well put. Additionally I would add that the if anything for me fiscal stimulus is not possible in Eurozone. Even this December we had comments like this comming from Jens Weidmann

    "Mr Weidmann considers the market to be the second disciplining mechanism: "Countries that do not conduct sound fiscal policies ought ultimately to pay higher financing costs." For this reason, he found market interventions such as the purchase of government bonds problematic."

    So Bundesbank (a very influential player in Eurozone monetary policy) is perfectly OK ditching inflation target if it can serve the good cause of disciplining Eurozone members for irresponsible fiscal policy. More fiscal stimulus means greater need for disciplining misbehaving Eurozone members.

    [Just aside, for some reason ECB is very fond of this sort of torturing mechanisms that produce self-propelling catastrophe. The other example being the "bad fiscal situation > higher VAT > higher inflation > tighter monetary policy > worse fiscal situation" spiral]

    There are other cases when monetary authority threw inflation target out the window for something else. One notable example may be bubble fears of influential bankers at Riskbank that forced Lars Svensson to resign in frustration.

    So to use Krugman's language, it is not lack of fiscal stimulus but "sadomonetarists" that are hurting the economy most and who actively not only prevent regime change, but also sabotage some of the more conventional tools available to central bankers (with QE being more an more conventional) You cannot fight this battle unless you have this crucial aspect sorted out.

  3. Nothing useful to add, but you are on a roll these days David. Liked your secular stagnation piece too, even though I don't really understand the "risk-premium" measure.

  4. Why is the fed targeting a 1-2% range of inflation? The timing is coincidentally since the start of QE, do they fear negative consequences from the unprecedented MB expansion. Maybe the risks aren't real and only perceived and the fed shouldn't worry. Even if risks aren't real heli drops would achieve 2% without needing to expand the MB to the levels seen under QE. Therefore there would be no perceived risks either.

    A regime change should include new policy tools.


  5. Elaborating on my comments on the previous post:

    1. When monetary stimulus comes in the form of QE, monetary tightening comes in the form of ending QE. However, if fiscal stimulus provided enough stimulus without QE, then monetary tightening would come in the form of raising rates. This would move us away from the ZLB. This is a good enough reason to do it by itself. Now is the time to think about what to do to get long-run interest rates up so we don't find ourselves as close to the ZLB next time a recession hits, and fiscal stimulus can help by providing excess stimulus for monetary policy to counteract.

    2. While the Fed's repeated QE tapering could reflect a desire to keep inflation below 2%, they could also reflect concerns entirely separate from the current inflation rate, most likely a concern that the cure of QE could be worse than the disease or just general discomfort with the use of QE due to the novel nature of it and the amount of money printing that has been required to make it work. So it's quite plausible that they see QE as a tool of last resort that they will do if inflation is far below target but not otherwise, but that doesn't mean if fiscal policy raises inflation to its target they will do whatever it takes to stop it.