Don't Trust the Inflation Data

(Bloomberg View) -- Federal Reserve Chairman Jerome Powell stated in testimony before Congress last week that policy makers will continue to strike a balance between avoiding an overheated economy and bringing the inflation rate back to 2 percent on a sustained basis. That's an interesting statement, as it seems to imply that an overheated economy will be measured in the standard inflation measures.

Yet even if the economy “overheats,” it's doubtful that the attendant demand-pull, cost-push factors that propel inflation will show up in the current measures of consumer prices. Indeed, the core inflation scorecard does not offer the same “inflation clarity” as it did in the past because it excludes valuable price signals from the asset markets and other areas of the economy. In other words, core consumer prices, or the ones that are targeted by policy makers, no longer show any consistent ties to the economic cycle, and the prices that matter more today are those that are not targeted, such as asset prices. 

Here's a case in point: In 2004 and 2005, the economy posted back-to-back relatively strong annual gains in real gross domestic product of 3.8 percent and 3.3 percent. The economy also faced considerable cost-push and demand-pull inflation, as prices for non-energy intermediate materials advanced 8 percent in 2004 and 4.5 percent in 2005, oil prices surged 35 percent and 37 percent in those two years, and average hourly earnings rose 2.5 percent and 3.2 percent. Demand-pull inflation was also evident in housing, with prices rising 10.3 percent in 2004 and 11.2 percent in 2005.

Potential drivers of consumer inflation were apparent, but how much actually showed up in reported core consumer inflation? Surprisingly, very little. According to the Bureau of Labor Statistics, core consumer prices rose 2.3 percent in 2004 and 2.1 percent in 2005. Why was inflation so benign and well below what one might have predicted? Although globalization and advances in technology no doubt played a part, one can't discount the changes in inflation “mechanics.”

Several years ago, government statisticians changed the measurement of housing costs, shifting it from a product (market prices) to a service (non-market rent). That triggered a material change in the composition of consumer prices, shifting the product/service mix from 54 percent/46 percent to 25 percent/75 percent, and reduced the sensitivity of core consumer prices to demand-pull factors, such as an increase in housing prices. House price inflation is no longer part of the traditional inflation measures, but it is clearly part of the inflation mantra policy makers should monitor as real asset prices are directly linked to money and credit conditions.

Another change involved the measurement of rents. Initially, the BLS surveyed both the rental market and owner housing market for rent data. Yet today it pulls all its rent prices from the smaller rental market. That change, which took place in the 1998, reversed the relationship between owner's rent and tenant rent. Now, owner's rent, which is not an actual price but what homeowners could conceivably rent their house for, runs at a slower pace than tenant rent, as has been doing so since this adjustment was implemented. That's a material change, since owner's rent account for a third of the core index.

Other mechanical shifts include changes for improved quality and product substitution, but BLS never provides any assessment of the scale of these adjustments so it is hard to say with any certainty how important these tend to be year after year.

No one is predicting the economy in 2018 will face the same scale and breadth of demand-pull and cost-push factors as it did in 2004 and 2005. But the recent trends in core inflation reveal how the mechanical changes, especially those that had to with housing prices and rents, create the impression that inflation is linear and slow-moving. In reality, though, some real assets that are no longer included in inflation calculations are moving higher more quickly.

Powell stated in his testimony that “careful judgments are required about the measurement” of variables used to help set monetary policy “as well as about the implications of the many issues these rules do not take into account.” This statement gives the impression that Powell will follow a more practical approach to monetary policy and not be guided solely by the news on core inflation.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Joseph G. Carson is the former global director of economic research at AllianceBernstein.

To contact the author of this story: Joseph G. Carson at jcarson21@aol.com.

To contact the editor responsible for this story: Robert Burgess at bburgess@bloomberg.net.

For more columns from Bloomberg View, visit http://www.bloomberg.com/view.

©2018 Bloomberg L.P.