« So What Happens in the Next Recession? | Main | Monetary tightening can lower real interest rates »

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

PLT is one of those ideas that looks good to monetarists, but I think most people -- especially those working in central bank offices, are naturally repulsed by it.

It forces the CB to be accountable for inflation misses. With the current regime, an inflation miss is forgotten as soon as we get back on target. So, if your point of view is that

1) we know very little about the causes of inflation, and that reducing excess inflation comes with terrible and asymmetric costs,
2) There is a "right" level of inflation in the economy, because inflation affects the economy

Then if the CB misses and prices go up by 20%, then a steep recession is created by the CB, and inflation gets back to target, do we need to impose a steep deflation just so we are "back on target"?

What if you change your model and assume that in k% of the periods, the CB screws up somehow and inflation becomes a brownian motion term. Then is PLT really a good idea, or is there an advantage to forgetting your mistakes? We don't really know a lot about inflation, so we should pick among policies with the understanding that all will fail regularly, and that these failures should not propagate weird shocks to the economy created only by the CB.

Finally, while the PLT cumulates past errors, it has a real credibility problems because you can always say "Well, next year we will start to make up for the 50% of cumulative misses. This year, we will just do what we think is right for the economy given it's present rate of inflation". So I don't believe in any of the supposed advantages in knowing future prices. It only adds more noise to future prices. Prices in 10 years might be anywhere from 30% lower (if the whole gap is made up) to 20% higher (if the gap is ignore) to anything in between. A k% miss in the past introduces k% uncertainty in future inflation.

You have less information about future prices with PLT than with IT as long as the CB has discretion on how to get back to target.

I remember in the last recession, the CB was supposedly promising higher future inflation to make up for past misses, and instead we got tightening as soon as the crisis was over. The reason we got tightening was the banks lost a huge amount of their money with low rates, thus there was incredibly pressure to get back to normal and normalize rates. It's really hard to convince the CB to tolerate a level of interest rates that is harmful to certain economic actors just to make up for past misses. Especially if those interests are banks. It was clear to anyone paying attention that the CB is not in fact going to tolerate more future inflation, and these are political constraints on the CB, given the relative power of banks and the importance of having a healthy banking system. It was all lies, and people are rational to form expectations the way they have.

Frank: I unpublished your comment. It's too far off-topic. Wait 2 days before commenting again, and keep it strictly on-topic.

rsj, You said:

"I remember in the last recession, the CB was supposedly promising higher future inflation to make up for past misses, and instead we got tightening as soon as the crisis was over."

That's interesting, and something I missed. By "the CB" do you mean the Bank of Canada? Do you have a link for that promise?

Nick, Interesting post. My view has always been "If the central bank makes a promise that it will adhere to, then people will believe it." Unfortunately my claim can't be tested with lab experiments.

With targeting NGDP, isn't it the case that you can't be sure what combination of real output growth and price level growth there is? What happens in a period of stagflation, for instance? There is negative real output growth and positive price level growth. Is that a desirable outcome?

rsj and Scott: I remember New Keynesian macroeconomists saying that central banks should commit to keeping interest rates "too low for too long" to help escape the ZLB for exactly that reason. But I don't remember any central bank actually making that commitment.

Scott: under the gold standard, is there evidence that prices were (in any sense) "mean reverting", or that people expected that high prices would mean prices would fall in future, or low prices would mean prices would rise in future?

Henry: if the economy is hit with supply shocks, then an NGDP target leads to smaller fluctuations in real GDP than a price level or inflation target. Think of a fluctuating SRAS curve, and compare RGDP fluctuations with a downward-sloping NGDP target AD curve to a horizontal inflation or price level target AD curve.

Scott, (and Nick)

I was talking about the Fed. The idea that a promise of higher future inflation can get us out of the lower bound goes back at least to Woodford, but it's a pretty obvious insight. It's in many of Bernanke's speeches. I agree that I don't remember reading this in FOMC minutes, but it was echoed by both Bernanke and Yellen several times in public speeches when they were governors, and also it was reflected for a while in policy -- FOMC annoncements kept extending the time out before increasing rates until about 2012, when they promised to keep rates at zero until the end of 2014 (this was made in January 2012). Thereafter, as the economy got better, we forgot about overshooting.

Try looking here:
https://www.ft.com/content/dcd7044a-904f-3fc0-9646-717546480713

"
Please use the sharing tools found via the email icon at the top of articles. Copying articles to share with others is a breach of FT.com T&Cs and Copyright Policy. Email licensing@ft.com to buy additional rights. Subscribers may share up to 10 or 20 articles per month using the gift article service. More information can be found at https://www.ft.com/tour.
https://www.ft.com/content/dcd7044a-904f-3fc0-9646-717546480713

The importance of this was that the Fed did not view the 2 per cent inflation target as a short-term ceiling. The central bank could justify aggressively easing monetary policy to get unemployment down, even if inflation moved temporarily above target. This thinking was then developed further by Charles Evans at the Chicago Fed, who suggested that the Fed should set an interim guidepost for the unemployment rate, subject to an inflation ceiling not too far away from 2 per cent. And, last month, Vice-Chair Janet Yellen indicated that optimal policy, responding to high unemployment rates, might allow a temporary overshoot in inflation in order to bring unemployment down more rapidly.
."

If you are willing to spend more time, you can find this in other public speeches by Fed Governors.

In terms of what a CB can commit to, there are some things that it can never credibly commit to, and policies are one of those things. Just imagine that in 2 years, we learn something new about inflation that makes PLT be a second best policy. Should we stick to PLT? Of course not. No institution would agree to that. Basic arguments about policy humility and robustness rule this out. Some promises are binding, others cannot be, and the public knows this. In order to have credibility, these future promises are either highly contingent, allowing escapes if we decide this isn't a good policy in the future, or they are time boxed to a few years at most. This necessarily blocks certain policies from ever being credible, and PLT falls into this category -- at least the infinite horizon symmetric version.

^^^Oops, I got copy and paste blocked from FT. Here is the quote I was looking for:

"The importance of this was that the Fed did not view the 2 per cent inflation target as a short-term ceiling. The central bank could justify aggressively easing monetary policy to get unemployment down, even if inflation moved temporarily above target. This thinking was then developed further by Charles Evans at the Chicago Fed, who suggested that the Fed should set an interim guidepost for the unemployment rate, subject to an inflation ceiling not too far away from 2 per cent. And, last month, Vice-Chair Janet Yellen indicated that optimal policy, responding to high unemployment rates, might allow a temporary overshoot in inflation in order to bring unemployment down more rapidly."

On the other hand, I think something like this would be credible:

In the next four years, we will target the price level by having an excessively loose policy if we undershoot our target, but we may elect to not make up for excess inflation if we overshoot our target.

But this policy, while at being possibly credible, is probably unappealing. But bottom line, the credibility of promise is not independent of what that promise is, and there is no reason to hope for a divine coincidence in which the set of credible promises intersect with the set of first best policies.

rsj: " But bottom line, the credibility of promise is not independent of what that promise is, and there is no reason to hope for a divine coincidence in which the set of credible promises intersect with the set of first best policies."

I think that's a useful way of putting it. But in the case of monetary policy and targeting inflation or the price level, we also need to remember: keeping a promise is easier if the promise is credible. Because any reasonable Phillips Curve says actual inflation depends a lot on expected inflation. The question that matters here is the amount of "inflation inertia". The Calvo Phillips Curve has no inflation inertia (which is unreasonable). The Taylor Phillips Curve (like the model in this paper, and the Bank of Canada's main models) does have inflation inertia. The costs vs benefits of PLT vs IT depend a lot on inflation inertia vs automatic stabiliser effects. Before the recession I was worried about inflation inertia more than automatic stabilisers. Now I think the right question to ask is: what is the best targeting horizon? Because inertia is only a problem if you try to change direction too quickly without warning.

And I keep trying to find a simple and intuitive way to do the math on overlapping price setting and inflation inertia vs automatic stabilisers. And failing. (The usual practice is just to throw all the equations into a computer and believe whatever answer the computer tells you. But I'm no good at that either, and anyway it's not my style.)

I'd be curious to see this experiment taking into account different types of subjects. People more invested in the markets such as business owners, bankers, investors probably play an outsize role in monetary transmission and they might also be better at predicting the effect of central bank moves if only because they have been naturally selected for this skill.

The benefits of PLT might be better than this research predicts.

There is no operational definition of inflation since various prices settle at a different pace. The central bank must select a representative sample that settles at the same pace as interest charges. Thus, no theory of price, just discretion.

I don't think this experiment is very useful to be honest. Is not like most economic decisions that are function of expected inflation are taken by normal people. We have a lot of people employed in forecasting inflation, and most decisions that depend on expected inflation use these forecasts. I believe we could get away with a much more complex system if we wanted too, and the lack of understanding of it by the public wouldn't make much difference.

Benoit and Arthur: I didn't notice in the paper who the subjects were. My guess is they were university students.

I can't make up my mind on whether it's the expectations of market professionals or ordinary people that matter most. But I think it's a bit of both. The key question here is households' and firms' consumption and investment decisions, including consumer durables. But what happens in financial markets might also matter, in affecting those decisions.

I very seriously doubt that households or firms -- in any meaningful amount -- form inflation expectations in a sophisticated way. The big firms don't hire staffs of economists, they turn to the market to hedge and borrow as needed for those specific input costs that concern them. The small firms read the wall street journal and get cranky that we are not on the gold standard (joke. kind of). Individual households mostly look at the past and what they see around them, in terms of inflation changes, and extrapolate a bit. Very few know or care about what policies the Fed is following, they care about what they see with their own eyes about the prices of things that affect them. It is all observation of prices, and extrapolate a bit. Which is also all that the CB is doing.

And I don't have a lot of confidence in wall street to discern price level targeting from other types of targeting. Because lets be honest, no one knows what they're doing in this space and no one understands the economy well enough to do much better than correctly guess the direction of change that will happen as a result of increasing or decreasing rates. When you are blundering in the dark, you take a step, see what happens, take another step, maybe if the coast has been clear for a while, you take bigger steps. When you hit a wall, you go in the opposite direction for a bit.

That is all that the central banks are doing. Well, that and they have researchers on staff that write some complicated math and have deep discussions while the policy decisions are still just a series of small steps in the dark.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Working...
Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.

Working...

Post a comment

Your Information

(Name and email address are required. Email address will not be displayed with the comment.)

Search this site

  • Google

    WWW
    worthwhile.typepad.com
Blog powered by Typepad