Barry Ritholtz is assigning Dan Gross to Jonestown, but Dan's New York Times article on why monetary policymakers focus on core inflation includes this bit that, in my opinion, hits the nail on the head:
... the Fed tends to focus on things that it can control. Not even a Fed chairman as powerful as Alan Greenspan can affect the price of oil by manipulating interest rates. "There's nothing the central bank can do about that, unless it figures out how to produce more oil," said Michael F. Bryan, vice president and economist at the Federal Reserve Bank of Cleveland.
But the Fed can control the amount of money circulating in the economy relative to the quantity of goods available. "So it tries to find the inflation signal common to all prices throughout the economy," Mr. Bryan said.
Thus considered, the core C.P.I. may be the best tool the Fed has to monitor long-term changes in prices.
Okay, as long as I am waving the hometown colors, this comes from the person at the Cleveland Fed who actually makes the decisions:
Let me spend just a moment explaining how I look at inflation. As a policymaker, I pay attention to the "core" measures of consumer price inflation as well as the overall "headline" measure of consumer price inflation, because the headline number can be somewhat misleading...
... the idea is to strip away the "noise" of volatile price changes, so that we can see the underlying inflation trend more clearly.
Of course, the FOMC's mandate is to control overall inflation: the average of all prices, including food and energy, as well as all the other things you buy. But we have found that measures of core inflation tend to be better predictors of future inflation than the headline rate, giving us a better picture of the true inflation trend.
I added the emphasis. Of course, Barry won't buy it -- and I mean really won't buy it-- but I think Mark Thoma did a pretty good job a few days back at laying out some casual evidence behind President Pianalto's assertion that core gives "us a better picture of the true inflation trend." I particularly liked this picture from Mark's post:
The fact that the difference between actual inflation and core inflation does not permanently drift away from zero is exactly the point: There is a lot of noise in the short-run, but over time core measures really do give you a sense of where things are eventually heading. (I am asserting here, of course, that it is mainly actual inflation that "catches up" with the core measure, not vice versa.)
New Economist left an interesting comment to Mark's post suggesting that, even though total and core rates of inflation do not diverge over time, the same cannot be said of their corresponding levels. Right:
This gets to the question of whether the central bank should have an inflation target or a price-level target. Some have argued that the latter may be preferable, at least in some circumstances -- here and here, for example. Others say inflation targeting is the better choice. Some say none of the above. Maybe this is all about to get more interesting.