Response to a comment by Waltraud Schelkle on my Fed2 post
My argument does not depend on there being a structural break in the relationship between headline inflation and the difference between non-core and core inflation - that is a break since about 2002 in the relationship shown in Chart 2 of the Fed2 post. It is rather about a break in the time-series process governing the relative price of non-core and core goods - to be precise, greater persistence, since about 2002, in the difference between the inflation rates of non-core and core goods. The fact that headline inflation tends to be higher when non-core inflation exceeds core inflation (that is, when the relative price of non-core goods is rising), need not imply any bias towards higher or lower headline inflation over time than would have been the case with a constant relative price of non-core to core goods. There would be no such bias if an increase in the relative price of non-core to core goods one period tended to be followed by a decrease of comparable magnitude during the following period(s). It so happens that the acceleration of globalisation at the beginning of the century has led to a rather long sequence of years in which an increase in the relative price of non-core goods to core goods has been followed by a further increase in the relative price of non-core goods to core goods. While this process of a rising relative price of non-core to core goods will not last forever, it is likely to have 'legs' for at least another decade or so (this medium-term pattern will of course be reversed temporarily if there is a cyclical downturn in global demand). So the structural break was in the relative price process of non-core and core goods - in Chart 1, not in Chart 2. Even though it is not essential to my argument, it is interesting to check whether there is visual evidence of a break in Chart 2 as well. Unfortunately, I haven't figured out yet how to colour the post-2002 observations in Chart 2 differently from the earlier ones, as you suggested. So instead I offer here two Charts (Chart 3 and Chart 4). Chart 3 has the 1987M1 to 2001M12 observations from Chart 2 of the Fed2 post, and Chart 4 has the 2002M1 to 2007M4 observations from Chart 2. It does indeed look as though the relationship has become somewhat stronger (i.e. the curve is steeper in Chart 4 than in Chart 3). That would be icing on my cake, but the cake would be pretty good even without it.