Boston Fed’s Public Policy Brief from May 2011 argues that changes in commodity prices do not affect long-run inflation rate despite warnings in the media.

In Do Commodity Price Spikes Cause Long-Term Inflation? the author, Geoffrey M.B. Tootell looks at the oil price in particular and says that, apart from its influence on wages in the 1970s, the oil price changes have had little effect on the core inflation changes.

Writes Tootell:

Neither theory nor evidence supports the notion that commodity price changes necessarily affect the long-run inflation rate. In the 1970s there appears to have been some effect on wages, but it is not evident in the latter sample period. Although commodity price inflation will affect total inflation in the short run, it does not appear to have affected core inflation, and thus total inflation, in the long run, at least since the 1970s.

Note in the graph below how the the energy CPI (black line) mimics the core (orange) during the 1970 – author calls this an “anomaly” -  but since the Volcker recession of early 1980s the energy line goes off on its own while the core change is going down – meaning that the price rises of the core are decreasing.

The key concept here is that the analysis is about trends in inflation changes and not an inflation or price level. In other words, price rises are assumed as a given but the rate at which they rise is what policy makers should be looking at.

“Central banks have very little control over relative prices in the long run. As a result, monetary authorities tend to target the longer-term trend in prices in general; they do not attempt to reverse shifts in relative prices,” explains the author, regrettably, only in the footnote.

The policy implication of this so-called “first derivative” (price changes) is that Fed should not be in business pricking any perceived commodity bubbles but making sure that these relative price changes do not get embedded in price expectations in the core.

The paper is available here.