Inflation and relative prices

There are persuasive reasons why we’d be better off today with an inflation rate higher than what we’ve seen over the last six months. But while a uniform expansion that raised all wages and prices by the same amount would be helpful, what the Fed could actually achieve in the present situation may be something less desirable.

When academic economists talk about inflation, we often think in terms of a single-good economy in which the concept refers unambiguously to an increase in the dollar price of that good. But in the real world, in any given month some prices rise and others fall, and we can only measure inflation in terms of the broad central tendency behind those individual price changes.

A recent research paper by Columbia Professor Ricardo Reis and Princeton Professor Mark Watson suggests that real-world measured inflation may behave very little like the textbook ideal. Reis and Watson introduce the hypothetical concept of a pure inflation shock as something that changes every price by x(t) percent, where in any given quarter t the magnitude x(t) is the same number for every item in the economy. Reis and Watson show how such a shock can be measured for any given quarter by observing the behavior of separate components of the PCE deflator. Their principal finding is that this concept of pure inflation in fact plays very little role in quarterly changes in broad price indexes such as the GDP deflator or the consumer price index, accounting for only 15-20% of measured inflation. The authors instead find that changes in relative prices are much more important than pure inflation for determining what happens to the broad CPI. For example, the measured deflation over the last 6 months is heavily influenced by falling energy prices.

If you think that the Federal Reserve is responsible for more than 15-20% of the variation in the CPI, the implication is that part of its influence comes from changes it causes in relative prices. But changes in relative prices– such as the huge run-up in energy prices in the first half of 2008– can be much more destabilizing than the textbook pure inflation.

I would therefore think that the Fed might be somewhat concerned by the surge in commodity prices over the last few weeks. The graph below plots the prices of 11 commodities since the Fed’s announcement of quantitative targets on March 18. Gold is the only one of these commodities that hasn’t gone up in price, with the average of these commodities up 13% over the last two months.


Some increase in relative commodity prices is certainly to be expected if we are indeed about to see a recovery in real economic activity. But this is a trend the Fed needs to watch closely from here, and could prove to be a significant limiting factor on how much the Fed can hope to achieve from monetary stimulus.

Because I for one do not think it’s a good idea to call for a replay of the 2008:H1 commodity market show.

Originally published at Econbrowser and reproduced here with the author’s permission.