Core vs. Headline Inflation

Recently, I was giving a speech and someone asked me what I thought about core inflation (which ignores food and energy) versus headline inflation (which includes everything, including food and energy).  I’ve been asked this question a lot and I don’t really have a problem with the idea of core inflation (but I understand that many people do).  I told him my position.  When he didn’t like that answer, I tried to make it better by offering him some free advice.  I said, “you’ll be fine – just don’t drive, don’t eat and most importantly, don’t drive to eat.”  Suffice it to say that this answer didn’t really help.  One more unsatisfied customer.  I think my work here is done.

Anyway, St. Louis Fed President James Bullard spoke on Wednesday about core inflation versus headline inflation (and he sees the issue differently than I do).  When you look at the chart below, you can see that headline inflation is more volatile:

consumer Pirce Index

President Bullard argued that there are four reasons why we use core inflation and that these reasons are wrong.  The reasons are:

1.     Headline inflation is more volatile than core inflation.

2.     The core predicts headline inflation.

3.     The “relative price” argument.  This says household budgets are fixed, so more money spent on one good must lower the amount spent on another good.

4.     Households may prefer that we focus on a subset of prices.

Bullard had many arguments as to why we should use headline inflation:

1.     The Fed risks losing credibility by ignoring inflation that households and businesses are aware of.

2.     The headline numbers were designed to be the best measures of inflation available.

3.     While the headline number is more volatile, monetary policy can be adjusted to use this.

4.     When policymakers use the core number, they make large changes in policy based on small changes in the core number.

5.     It would be easy to smooth the headline number.  One simple way is to measure the headline inflation for a one-year period (rather than annualizing a one month change).

6.     It is silly to build a model that simply says that core inflation predicts headline inflation.  We need to include other factors such as expected inflation, developments in the real economy, the stance of monetary policy and other variables.  The point is that it’s wrong to say that core inflation (alone) predicts headline inflation.

7.     There is no convincing evidence that a model with core inflation adds special information than other (more full) models.

8.     The idea that core inflation predicts headline inflation is unclear.  You have to decide what the forecast period is, what function of core inflation to use and the measures of core and headline inflation to use.

9.     A trimmed mean inflation measure (such as one calculated by the Dallas Fed, where they exclude the highest and lowest price changes each month) does better than core inflation (where we always exclude the same two factors).

10. If you think of the household budget as fixed, a change in relative price will lead to a change in consumption of that product.  Relative price movements are already accounted for in the existing indexes.  Therefore, when the entire price index rises, it really does mean that there is inflation in the economy.

11.  The Fed does not control any one element, but only has some general control over the entire price index.

12. If people spend more on energy (for example), they spend less on other items and this puts downward pressure on those items.  By ignoring food and energy (when it rises), we understate the true inflation rate.

13. If global demand of energy continues to increase over the long term, prices will increase and we will systematically understate inflation (if we ignore it).

14. You could argue that we care about sticky prices (which do not quickly adjust to supply and demand) and we don’t care about fully flexible prices like oil.  Similarly, we could argue that we focus on the sticky prices in our country and ignore the flexible import prices.  But, these are very academic arguments and not commonly discussed in the real world.

I really like President Bullard and his thoughts, but I actually believe core inflation is important.  I realize that this is probably a minority view and that most people get really frustrated by the idea of “core inflation,” but it scares me to think of making policy based on a volatile headline number.  In addition, food and energy costs are outside of our control in the short term.  To me, the real fear (and what we’re watching) is to see when food and energy inflation has worked its way into the cost of other goods (core inflation).