Core For: More

Barry Ritholtz replies in the comments section of my earlier post about core inflation:

By framing the issue the way you did, you get one answer.

I prefer a different set of questions. These of course generate a different set of answers.

My questions:

1) What is the actual rate of inflation?

2) Why does the BLS model (the official inflation rate) vary so greatly from the real world experience?

3) What are the Fed policy repercussions of the spread between the two?

4) What does this mean to consumers? Investors? Savers?

I am not surprised that traditional economists have circled the wagons around my attack on the credibility of BLS and the Fed. Thats what all Guilds do when they sense a challenge to their authority . . .

He has more on his own blog, but let me try my answers. I first note that these questions, as he stated them originally, mostly relate to the full CPI, not particularly to the core CPI. The core is not “the official inflation rate,” nor should it be. At least in my view, the “official inflation rate” should measure what prices in general have done retrospectively; it gives us information about the past, and the core would be the wrong information. To the questions,

1. What is the actual rate of inflation?
   
   It depends on what you mean, specifically, precisely, operationally, by inflation. It is a mistake to think there is one “true” inflation rate, because different prices are changing by different amounts and in different directions, and because quality is changing in ways that affect different people and different businesses differently, and because the effects of these changes – even on an individual – are often impossible to measure. How can we know, for example, how much it is worth to have a more powerful computer for the same price? In most cases, all we can do is make an educated guess, and depending on how one chooses to educate the guess, many different reasonable answers are possible.
   
   It is also a mistake to go (as Barry Ritholtz seems to do in the post cited) directly from the premise that “inflation is primarily a monetary phenomenon” to the conclusion that inflation should show a consistent quantitative link with a particular measure of the money stock. The relation between inflation and (any particular definition of) money depends on the evolution of payments technology and potential output. In particular the finding that the CPI diverged from M2 and M3 starting in the mid-90s tells us very little: either productivity really did start growing more quickly (in which case we should expect such a divergence) or else it didn’t.
   
   
2. Why does the BLS model (the official inflation rate) vary so greatly from the real world experience?
   
   The question is not very meaningful unless you can specify what you mean by “real world experience” and demonstrate that it differs from the BLS model. If you mean subjective “real world experience,” then I’m inclined to blame psychology rather than measurement for the difference.
   
   
3. What are the Fed policy repercussions of the spread between the two?
   
   OK, never mind. On his own blog, Barry changes the question:

   Why does the Fed Focus on the Core rate, and not the actual rate? What are the Fed policy repercussions of this?

   To the extent that the Fed does focus on the core rate, it does so primarily for two reasons (and more which I may discuss in a later post):
   
   
   • Using short-run (e.g, 1 year or less) measurements, the core rate has generally proven to be a better predictor of future inflation than the full rate.
     
     
   • The core represents items with relatively sticky prices, so price pressures on the core do more damage than price pressures on noncore items. In particular, downward price pressure on the core causes recessions. Because noncore prices are more flexible, the damage is absorbed by prices before it can cause distortions in quantities.
   
   The repercussion, over the past 5 years of rising energy prices, is that we have had 5 years of economic growth when we could have had an ongoing recession.
   
   
4. What does this mean to consumers? Investors? Savers?
   
   To consumers, it means more of them have jobs than otherwise would. It also means that the decline in their real incomes has come in the form of price increases rather than wage cuts, so they can (wrongly) blame it on the Fed rather than the scarcity of oil. (It is always nice to have an institution to blame instead of an abstract concept.)
   
   To investors, it means that, in retrospect, they should have owned TIPS instead of nominal bonds. By the way, the Treasury is still selling TIPS, if you want ’em. Unfortunately, just as in 2002, we lack perfect foresight as to what the noncore component of inflation will do. It may go down, and you may get screwed owning TIPS. But if you want a secure real return, it’s available.
   
   To savers, I’m not sure what it means. Barry has a wee bit of a point that, by focusing on the core without making sufficient qualifications, the Fed may be misleading people into thinking that the core is the actual inflation rate, and savers will be disappointed when it comes time to spend their savings. I don’t see this as a reason to de-emphasize the core for policy purposes but as a reason to be more careful when speaking about it.

If traditional economists are circling the wagons around the core, that’s because it really is a better policy target. (As to BLS methods, that’s another question entirely, about which Barry and I might be able to find some common ground...but this post is already too long.)