Comments on economics, mystery fiction, drama, and art.

Thursday, April 14, 2011

Do we need to worry about whether the relationship between the core rate of inflation and the headline rate of inflation has changed?

Well, that's a long title for a blog post.

James Bullard, the President of the St. Louis Federal Reserve Bank, writing in the Bank's The Regional Economist, suggests that we might have reason to be worrying about reliance on core measures of inflation:

Since June 2010, the two measures have been diverging slowly, with core inflation below headline...during times of continuous increases in the relative price of energy, perhaps core PCE [inflation--DAC] is a misleading indicator of underlying inflation trends. This implies that core PCE may not be a good predictor of future headline inflation after all. Under these circumstances, headline PCE inflation should probably have more weight in policymaning decisions than core PCE inflation.

So I decided to take a look; unfortunately, I continue to have trouble pasting Excel charts into Blogspot, so I'll just have to relate what I see when I graph core inflation and headline inflation. (I calculated the rate of inflation as the percentage change in the monthly all-item or core PCE from 12 months earlier.)

And I'm not seeing the problem. The headline rate of inflation, as measured by the all-item PCE, has become more volative since about 2005. And that volatility suggests that food and energy prices have also become more volatile recently. It is accurate to say that inflation measured by the all-item PCE did in fact exceed inflation as measured by the core PCE early in 2010 (when oil prices were rising), but the opposite was true in 2009 (when oil prices were falling). Isn't this what we'd expect? Wouldn't we be surprised if this did not happen? (Incidentally, I also looked at this using the all-item and core CPI, and the results are the same.)

What Mr. Bullard suggests might be a reason for decreasing our reliance on inflation measured by core price indexes to forecast future headline inflation is in fact a feature, not a bug, of using core measures of inflation. For now, I see nothing to be concerned about.