If I seem to be “picking” on Scott Sumner this week it’s because he’s been passing around information which I find highly interesting over the last few days (in addition, one can almost never go wrong by following Scott’s thought leadership – although I can imaging him saying “get your own ideas!”).

In his latest, Scott posted a reference to a series of blog postings by Mike Bryan from the Atlanta Fed where he discusses alternate ways of measuring CPI data, and describes the communication problem presented given that the public considers the definition of inflation in an entirely different way than monetary policymakers do. I agree that it is a problem simply because within the makeup of price changes there are a combination of forces impacting prices besides those resulting from monetary policy.

The post that Scott commented on was the last in a series of three posts that were a rough translation of the talk given by Bryan at the Cleveland Fed. When I followed the link, I read the entire series that cover alternate measures of inflation based on the CPI, one being a mean measure to reduce the monthly volatility, and another that reweights the CPI toward prices that are, in his judgment, more likely to display the common inflationary thread that could be said to result from monetary inflation.

These are all great ideas, and I don’t intend to down play their importance when the meaning of the CPI has been misconstrued even by policymakers to a point of monetary policy becoming more about politics than economics, hurting the people they are supposed to be serving. Even though I am about to criticize the entire idea of measuring inflation, and how much importance it should have in the formation of monetary policy, I cannot overstate the importance of what appears to me as an effort at level-setting regarding the topic in order to try and restore sanity to the conduct of monetary policy in general.

During his talk, Bryan presents two different versions of the CPI that are different from what we refer to as “core.” His reasoning behind a dismissal of the core price index is that it is arbitrary, and the prices of these items impact the prices of most other goods. Here’s what he says about it:

The Economist retells a conversation with Stephen Roach, who in the 1970s worked for the Federal Reserve under Chairman Arthur Burns. Roach remembers that when oil prices surged around 1973, Burns asked Federal Reserve Board economists to strip those prices out of the CPI “to get a less distorted measure. When food prices then rose sharply, they stripped those out too—followed by used cars, children’s toys, jewellery, housing and so on, until around half of the CPI basket was excluded because it was supposedly ‘distorted'” by forces outside the control of the central bank. The story goes on to say that, at least in part because of these actions, the Fed failed to spot the breadth of the inflationary threat of the 1970s.

I have a similar story. I remember a morning in 1991 at a meeting of the Federal Reserve Bank of Cleveland’s board of directors. I was welcomed to the lectern with, “Now it’s time to see what Mike is going to throw out of the CPI this month.” It was an uncomfortable moment for me that had a lasting influence. It was my motivation for constructing the Cleveland Fed’s median CPI.

I am a reasonably skilled reader of a monthly CPI release. And since I approached each monthly report with a pretty clear idea of what the actual rate of inflation was, it was always pretty easy for me to look across the items in the CPI market basket and identify any offending—or “distorted”—price change. Stripping these items from the price statistic revealed the truth—and confirmed that I was right all along about the actual rate of inflation.

This is a rather long quote that can be summarized by saying that in the past there was a habit of throwing prices out of the CPI index as being distorted, though after years of analyzing the monthly CPI report, he became skilled at finding the distortions. I suppose it was intuitive from looking at the bars of the components. It’s great to have a sort of a CPI interpreter on that journey of trying to find that missing link between monetary distortions from other factors. He even goes through his reasoning, and processes developed with others in establishing at least two other ways to measure inflation that are much less “crude” than core.

The problem I see here is not that his work is not useful. It may be quite useful, and is likely an improvement over the current basis of policy. But I am not convinced that even these are the “silver bullet,” the sure-fire, fool-proof way of determining the inflation rate based only on monetary factors because his alternates do not agree with one another, and neither indicated an issue in the 2009-10 timeframe that would have indicted a need for loose policy. I am not sure that throwing out the tails in the mean measure or reweighting data contained in the CPI is different in effect with respect to arbitrariness from throwing out other price data in core. It may be a more sophisticated way of throwing things out, with some more pointed rationale attached, but there is still very much a question in my mind as to whether or how much these alternate measures capture the essence of price movements as a result of monetary policy. I don’t think they do, and even this cannot solve the problem of the Fed simply not knowing the difference and responding in an incoherent manner to the conditions on the ground.

At this point, my post can follow a couple different lines of thought. The first being that with a low target, the margin of error in conduct of policy is condensed and the problem of capturing the portion of price movements that are related to the conduct of policy is more likely to have a much greater impact. Making a mistake will cause more damage than benefit to be gained.

The other line of thought is more toward the practical, like whether basing the nominal anchor for monetary policy on a phenomena that simply cannot be reliably measured nor extrapolated from existing data is a rational choice. There is simply a whole raft of problems with the current policy framework that must be worked through in order to create a livable solution. There is little doubt that the solution would be quite complex with plenty of fluid exceptions, and plenty of excuses to be thrown around when the inevitable mistake occurs. It just not appear to me to be a definition of sanity or anything resembling rational or even prudent conduct given the risks of throwing the baby out with the bathwater that are posed nearly every day.

While there may be current issues with measurements of NGDP, those are far more solvable than this; and I’m left scratching my head regarding the stubbornness of monetary officials to even given NGDPLT a fair hearing. We desperately need a regime change, or at the very least, a coherent explanation was to why IT is better given all these measurement issues and political problems that come attached.

Advertisements