The Fallacy of the ‘Core’ Rate of Inflationby
Stefan M.I. Karlsson
In recent years, the news media have increasingly focused on the so-called core (excluding food and energy) measure of inflation. Often we hear how the all-items index may have been higher than expected, but we should still cheer because the core number was lower than expected. The reasoning behind this is that food and energy prices are supposedly more volatile than other components, something that is certainly true in the case of energy but less true of food, which in the aggregate is hardly more volatile than some "core" categories. Individual sub-categories in food like vegetables may be highly volatile but that is usually counteracted by other sub-categories remaining stable or even moving in the opposite direction. The reasoning for excluding food and energy is again their real or alleged volatility, which supposedly make them unreliable as inflation indicators. But the idea that the core index in any way better measures "underlying" inflation pressures should be rejected for several reasons. Firstly because the exclusion of certain items contradicts the justification that pro-government economists use for focusing on price changes rather than changes in the money supply. Supposedly focusing on prices is better because that shows how much the purchasing power of money has actually changed rather than just focusing on money supply changes which only on a long-term basis can be expected to fully influence prices of goods and services and even then only adjusted for the change in the supply of goods and services. But if focus should be on the prices people actually face rather than underlying long-term trends then it makes no sense to exclude food and energy which people in the real world certainly cannot exclude. And if focus should be on the underlying long-term trend of inflation then focus should be on money supply changes in relation to changes in the supply of goods and services. Whatever principle you use, focus on core price indexes cannot be justified. Secondly because it is far from evident that volatile price changes are worse indicators of underlying price trends than more inflexible prices. Often the opposite is true as a money supply increase for example is likely to have a quicker impact on prices which change quickly than prices which change only rarely. Thirdly because an increase in energy prices based on a autonomous factor like reduced oil supply will likely have a deflationary effect on other prices because it reduces people's purchasing power meaning that underlying price inflation will be underestimated by the core index when energy prices rises. This effect is enhanced in the statistical numbers by the fact that when fuel prices rise while rents have in the short-term not been adjusted upwards then the "home owners equivalent rent" will be decreased as it is calculated based on rents minus fuel costs. Meaning that any increase in fuel costs for home owners will be exactly matched by an equivalent decrease in "home owners equivalent rent" which in turn means that higher energy prices will at least in the short term have the effect of lowering the core inflation numbers. Sometimes it is also argued that oil prices could rise on a autonomous factor like reduced supply or increased demand from China or other non-American sources and since this increase then was not the result of US monetary policy it should not influence it. But if we accept the view that price increases based on reduced supply of a certain good is not real inflation then we should also regard price declines or lower-price increased based on higher supply of goods as real inflation. In which case focus should be on money supply and not price indexes, including core price indexes. Again we see how the argument against the all-items price index can be turned against the core price index. Some people might now wonder why the core index is so useful for governments to deceive people into believing inflation is lower than it is. After all, energy prices do not always rise faster than "core" prices. During some time periods like the mid 1980s and the late 1990s energy prices fell sharply and the "core" inflation reading was then higher than the all items inflation reading. That is true, but governments can still use the core index and make sure to focus on it during all times when energy prices rise. But during the periods when energy prices fall, you will see governments and its apologists focus on the all-items index. So with the existence of both an all-items and core measure governments can simply focus on whatever index shows the lowest increase. Shifting method of measurement is a common tactic among those who want the Fed to accelerate monetary inflation or at least not decelerate it. Larry Kudlow and other supply-side economists in 1998 and 1999 for example used the CRB commodity price index and the gold price as evidence for the absurd assertion that the Fed then -during the height of the tech stock bubble- pursued a "deflationary" monetary policy. Both the high money supply growth and the low but still positive increases in consumer price indices were dismissed as irrelevant. But now that the prices of gold and other commodities have skyrocketed (the CRB index is up more than 15% in a year) they are suddenly deemed irrelevant with the core personal consumption expenditure (PCE) deflator index (chosen over the CPI since it increases slightly slower) instead now being considered the important measure as it shows only a slow increase at about 2% a year. Similarly in a January 1999 speech Fed Governor Roger Ferguson did not regard the core measure as worthy of mentioning and accepted the then lower oil prices as a factor which has held down inflation. While he did warn (correctly) that the oil price decline would not last, he also warned that several factors influencing core measures, like falling non-energy commodities and a rising dollar, would not last either and he did not make any distinction between core prices and energy prices. Now however, the Fed like the supply-siders, have declared that they focus on the core PCE deflator and regard energy prices as irrelevant. How convenient. August 5, 2005 Stefan M.I. Karlsson [send him mail] is an economist working in Sweden. |