The Inflation Update

As the reports of declining prices accumulate, talk of deflation is intensifying. The advance GDP report for the third quarter of 2001 showed a decline in virtually every price index. It was only the idiosyncratic treatment of reinsurance payments for the World Trade Center attach made by foreigners (an import) that kept the overall GDP chained price index positive. If we were really poised on the verge of a prolonged deflation, that would be cause very worrisome. My view is that we are not, and let me explain why.

The most recent CPI release showed headline inflation at a -4.0% at an annual rate (a..r.) for the month ending October. By contrast, core measures showed healthy increases, with the CPI excluding food and energy up 1.9% (a.r.) and the Median CPI of the Federal Reserve Bank of Cleveland rising 5.1%, its largest one-month increase since February 1994. The more representative 12-month changes showed increases of 2.1% for the headline CPI, 2.6% for the CPI excluding Food & Energy, and 3.9% for the Median CPI.

Divergences of this magnitude are unusual in a set of inflation measures that are all based on the same underlying data. To understand the deviation of the traditional core from headline inflation, we need look no farther than the decline in the price of energy. Motor fuel was the most extreme, as its price fell by more than 10% for the month, or just under 75% at an annual rate. It is always easier to hold a cartel together when revenues are high. As the world economy recovers, and I think it will, we can expect high oil prices to return. But in the meantime, we can enjoy the boost that this is giving to growth.

This brings us to the much more puzzling behavior of the Median CPI and its out-sized increases over the past year. Regular readers of this update know that the shelter plays an important role in the calculation of the Median CPI, and is central to my views of inflation trends. Increases in the estimated price of housing services, both rent and owner-occupied, has been high and rising for some time now. Owner equivalent rent (OER) rose 5.3% (a.r.) last month, and is up 4.2% for the past year. This way above average, and is the reason the Median CPI show inflation so much above the other measures. But there is reason to believe that this is an overestimate by as much as one percentage point.[1] Even so, that still leaves us with an estimate of the inflation trend in excess of 3% per year – and that’s my current view.

Looking further at the detail of the release, we see further ominous signs in the 5.4% increase in medical commodities and a 5.7% increase in medical care services. Overall medical care prices have now risen 4.6% for the past year, and appear to be accelerating. Further evidence that service price inflation has not fallen, comes from the 3.5% (a.r.) increase in food away from home this past month, and 3.1% for the past year.

This all leads me to two conclusions. First, I see no signs of a broad-based deflation. In fact, it is quite the opposite. Inflation seems to be slowly on the rise and now stands close to 3%. Second, we need to change our perspective somewhat as for the first time in decades we are living in a world in which policy has delivered very low average inflation. When the average is low, there will always be some prices that are falling as a consequence of some combination of weak demand and strong supply. The sellers of these goods and services will cry that there is deflation. Deflation is disastrous, they will shout, and so something has to be done about it immediately. It is always bad news when you learn that you are below average. But policymakers need to be concerned with trends in aggregate inflation, not the misfortunes that have befallen individuals. And today, the average is rising!

Consumer Price Inflation, Various Measures
(Through October 2001, all data at an annual rate)

Detail for Computation of the Median CPI
October 2001
Component	Annualized 1-month % change	Relative Importance	Cumulative Relative Importance
Motor fuel	-74.7	3.4	3.4
Fuel oil and other fuels	-41.5	0.3	3.7
Tobacco and smoking products	-32.1	1.4	5.1
Gas (piped) and electricity	-25.8	3.9	9.0
Public transportation	-21.2	1.4	10.4
Lodging away from home	-20.0	2.3	12.7
Women's and girls' apparel	-4.9	1.7	14.4
Used cars and trucks	-3.8	1.8	16.3
Personal care products	-0.8	0.7	17.0
Fresh fruits and vegetables	0.5	1.1	18.1
Cereals and bakery products	0.6	1.5	19.7
Household furnishings and operations	0.9	4.5	24.2
Recreation	1.1	5.9	30.1
New vehicles	1.7	4.6	34.7
Tenants' and household insurance	2.3	0.4	35.1
Car and truck rental	2.4	0.1	35.2
Personal care services	2.6	1.0	36.2
Miscellaneous personal services	3.2	1.5	37.7
Water and sewer and trash collection services	3.3	0.9	38.7
Motor vehicle fees	3.3	0.6	39.3
Motor vehicle parts and equipment	3.5	0.5	39.8
Food away from home	3.5	5.7	45.6
Footwear	4.0	0.8	46.4
Alcoholic beverages	4.1	1.0	47.3
Rent of primary residence	5.1	7.2	54.6
Owners' equivalent rent of primary residence	5.3	20.9	75.5
Motor vehicle maintenance and repair	5.3	1.6	77.1
Medical care commodities	5.4	1.3	78.4
Processed fruits and vegetables	5.6	0.3	78.7
Medical care services	5.7	4.7	83.4
Motor vehicle insurance	6.4	2.5	85.9
Miscellaneous personal goods	6.5	0.2	86.1
Communication	6.6	2.5	88.6
Nonalcoholic beverages and beverage materials	7.1	1.0	89.6
Education	7.2	2.9	92.5
Meats, poultry, fish, and eggs	8.5	2.6	95.2
Other food at home	8.5	2.0	97.1
Dairy and related products	10.4	1.1	98.2
Men's and boys' apparel	16.6	1.2	99.4
Infants' and toddlers' apparel	22.4	0.2	99.7
Jewelry and watches	36.7	0.3	100.0

[1] In the May 2000 update I noted that research by Richard Peach and Jonathan McCarthy at the Federal Reserve Bank of New York suggested that the owner-equivalent rent portion of the CPI had been understated in latter half of the 1990s. Their conclusion was based on a combination of cyclical and demographic factors. What we are seeing now may be a reversal, and so OER may be somewhat overstated.

Detail for Computation of the Median CPI