By: Phil Kernen

In July 2022, the annual increase in the Consumer Price Index (CPI) was 8.5%. Excepting the 9.1% advance in June, this matches the highest since 1981, only then it was declining after recording three straight years of double-digit price increases from 1979 to 1981. Is it possible we will see double-digit inflation again in 2022?

We have written before about the number of ways to calculate inflation. Various branches of the Federal Reserve produce dozens of them, even ignoring adjusted calculations of each that exclude energy and food.  Consider a sampling:

Atlanta determines whether an inflation input falls into one of two categories: sticky items whose prices change more slowly like tenant rent, owner’s equivalent rent, and medical care services, and flexible items whose prices change more rapidly like new and used vehicles, groceries and apparel. Sticky inflation for July increased 5.2%, and flexible inflation increased 16.3%.

Cleveland computes a Median CPI whose expenditure weight is in the 50th percentile of price changes. A separate trimmed-mean excludes extremes by ignoring components at the top and bottom of the distribution. Median CPI for July was 6.3%, and the Trimmed Mean was 7.0%

Dallas defines a trimmed mean too, but takes it further, excluding more results to the upside than the downside which usually leaves it lower than other inflation measures. They apply their method to the favored inflation metric used by the Fed, Personal Consumer Expenditure (PCE) and recorded a 4.3% increase for June. 

New York calculates a new Underlying Inflation Gauge which uses a host of different inputs, including other inflation measures. The UIG for July was 4.7%.

There is no single, correct way to determine inflation. There are too many goods and services, substitution choices, weights, and geographic differences to consider. The point is that every calculation shows that inflation is running hot and significantly exceeding price stability targets.

Even the standard-bearer CPI, now four times higher than the official inflation target set by the Federal Reserve in 2012 and the most-commonly referenced inflation measure used for government programs and private contracts, has undergone six comprehensive revisions over several decades and numerous smaller additional improvements.  

One notable change occurred in 1980 to introduce the rental equivalency concept. Before the update, changes in housing prices were direct inputs to CPI. Following the change, housing costs were indirectly input, determined by imputing what monthly rental payments a homeowner would incur to live in their home. As a result, changes in housing prices worked their way into CPI slowly as higher housing prices translated into higher rents. CPI calculated using the prior methods would reflect a 15% increase following the strong run-up in housing prices over the last few years.  

If today’s CPI calculation reaches double-digits, we can blame shelter, energy, and wages. We addressed shelter in an earlier article, talking about how the Dallas Fed estimates that changes in housing prices find their way into CPI shelter costs 18-24 months later. Housing price growth accelerated in 2020-21 and shelter costs are rising now – CPI shelter is up 5.7%, and its monthly changes have been slowly accelerating. 

Oil has fluctuated around the $100 threshold for a barrel, while demand has reached, and will exceed, pre-COVID levels. Meanwhile, OPEC production isn’t increasing, American producers invest only enough to maintain output, and the ability for Russia to deliver oil under a cloud of new sanctions has become highly uncertain. And wages are growing as employees press employers for pay raises to keep up with general inflation pressures. Employers respond by raising consumer prices to match rising costs, supporting the wage-price spiral pushing both towards an uncertain end. 

The Fed is sure that inflation will moderate by year-end and are finally responding. The data suggest inflation is sticky, and the pieces are in place for inflation to move higher.