Will the True PPI Stand Up?

You have to be careful in determining which series you’re looking at. Here are four measures of m/m PPI inflation (latest release, discussed here).

Figure 1: Month-on-month inflation (not annualized) for PPI for final demand (blue) [FRED series PPIFIS], PPI for all commodities (brown), PPI for final demand, finished goods (green), PPI for final demand excluding food and energy (red). NBER defined recession dates peak-to-trough shaded gray. Source: BLS, NBER, and author’s calculations.

FRED series PPIACO is a series that extends back to 1913. For those of us who went to learned about the PPI in the 1980s and 1990’s, this series is what was referred to — a goods only measure. In exchange rate determination papers I wrote in the 1990’s, for instance, I would use the PPI as a proxy measure for tradables goods prices, and CPI for nontradables.

One important aspect of this PPI measure is that it includes prices at different stages of production; hence there is multiple counting of the same product at different stages (e.g., intermediate vs. final).

Core measures are not straightforwardly identified, particularly if one is averse to reading the explanatory notes (or just averse to reading in general). When looking at indices excluding energy and food, there are (at least) a couple of measures that behave quite differently.

Figure 2: Month-on-month inflation (not annualized) for PPI for final demand excluding food and energy (blue) [FRED series PPIFES], PPI for final demand finished goods excluding food and energy (pink). NBER defined recession dates peak-to-trough shaded gray. Source: BLS, NBER, and author’s calculations.

So, be sure you know which PPI you’re looking at (and that other people are referring to). FRED series PPIFIS and PPIFES (final demand and final demand ex-energy and food) are the most commonly cited (a hint is that the folks running FRED have given specific mnemonics to them, and Bloomberg surveys reference these two series).