Paul Krugman explains why Keynesian economists have now begun to switch to a fifth definition of inflation, “core inflation”, that is distinct and increasingly distant from the Keynesian school’s theoretical, official, textbook, and practical definitions.
Since I’m taking a break from shoveling, I thought I might take a few minutes to address an issue that seems to confuse many people: the idea of core inflation. Why do we need such a concept, and how should it be measured? So: core inflation is usually measured by taking food and energy out of the price index; but there are alternative measures, like trimmed-mean and median inflation, which are getting increasing attention.
First, let me clear up a couple of misconceptions. Core inflation is not used for things like calculating cost-of-living adjustments for Social Security; those use the regular CPI. And people who say things like “That’s a stupid concept — people have to spend money on food and gas, so they should be in your inflation measures” are missing the point. Core inflation isn’t supposed to measure the cost of living, it’s supposed to measure something else: inflation inertia….
The standard [core inflation] measure tries to do this by excluding the obviously non-inertial prices: food and energy. But are they the whole story? Of course not — and standard core measures have been behaving a bit erratically lately. Hence the growing preference among many economists for measures like medians and trimmed means, which exclude prices that move by a lot in any given month, presumably therefore isolating the prices that move sluggishly, which is what we want.
And then these great minds of the dismal science wonder why their models don’t work in either a predictive or an explanatory manner. Of course, the reason they need to keep coming up with these additional “definitions” is because not because the “measures have been behaving a bit erratically” but because the conclusions that are based on the definitions don’t line up with what is visibly taking place in the actual economy.
The red line is the standard “core inflation” CPI-FLENS measure that Krugman is citing in his second “Core Logic” post; the blue line is the standard CPI-U. However CPI-FLENS cannot possibly measure the inflation inertia because the practical application of the various Keynesian definitions of inflation requires something to measure those price changes against. Simply removing some of the more volatile (and generally higher) prices from the price index only smooths out the price index while tending to reduce it. Since inflation – and therefore its inertia – depends upon the comparison of those price changes with changes in the production of goods and services, (in practical terms, with GDP minus the price increase of one’s preferred index, be it CPI-U or CPI-FLENS), anything that leaves GDP out of the equation cannot be considered the inflation inertia anymore than the mere change in prices can be considered inflation if it does not take the change in production into account.