Inflation Is Sticker Than It Looks. Consider the Cleveland Fed’s Data

Text size

Inflation is sticky — and even stickier than it looks.

Investors had better get used to seeing that word. Sticky inflation is underlying inflation, or inflation in areas where prices tend to change relatively slowly. Consider it the opposite of more volatile inflation in categories such as food and energy, which economists and policy makers back out of inflation readings to get to what they call core inflation.

After the Labor Department reported an 8.3% year-over-year increase in the total consumer price index for August, a bigger gain than expected, economists and Wall Street strategists agreed that the latest data show inflation is sticky. Michael Ashton, investment manager at Enduring Investments, notes that about 70% of the CPI rose at an annual pace of more than 6%. 

Barron’s emphasized the sticky-inflation problem last month, after investors and economists welcomed a slight slowdown in consumer prices during July. Many believed so-called peak inflation had finally been reached. But as Jefferies chief economist Aneta Markowska put it Tuesday, the latest CPI report puts the “peak inflation” assumption into question and shows that the labor market and demand -– not supply — problems are driving price increases.

And, as former Treasury Secretary Larry Summers tweeted, “core inflation is higher this month than for the quarter, higher this quarter than last quarter, higher this half of the year than the previous one, and higher last year than the previous one.”

But the situation is even worse than that. For all the attention that the government’s monthly CPI release attracts, there are better numbers for investors to watch now. The Federal Reserve Bank of Cleveland puts its own spin on the government’s CPI every month, in an effort to capture underlying inflation by dropping extreme increases and decreases. The bank’s CPI data suggest underlying inflation is even stronger than the August CPI report reflects.

The two key gauges from the Cleveland Fed are the median CPI and the trimmed-mean CPI. The former represents the one-month inflation rate of the component whose expenditure weight is in the 50th percentile of price changes. The latter is the weighted average of one-month inflation rates of components above the eighth and below the 92nd percentiles. In August, the median CPI rose 6.7% from a year earlier, marking the highest level ever in a data series dating back to 1983. The trimmed-mean CPI increased 7.2% from a year earlier. 

Of all the inflation data reported Tuesday, Harvard economics professor Jason Furman says he prefers the Cleveland Fed’s median CPI. He calls the August print “extremely ugly,” and notes that looking at three-month changes shows how sticky inflation is firming up, even as core prices–- or those excluding food and energy–- are coming down. One month doesn’t make a trend but three months do, and the trend is clear. 

The CPI report alone means the Federal Reserve is going to raise interest rates by another 0.75% this month, and any remaining hope of a smaller rate hike is dead. Traders on Tuesday started pricing in a chance of a full-point hike, with

CME

data showing that probability at 18%, up from zero a day earlier. 

More important than whether the Fed surprises with a one percentage point increase next week is how high rates will ultimately go. The thinking has been that the Fed is front-loading rate increases, resulting in a so-called terminal rate of about 4%. But Markowska at Jefferies calls the August consumer price data a game changer, and now expects that the Fed will hike at least to 4.5%.

Markowska thinks the Fed will lift rates by 0.75% this month and at its next policy meeting, in November. That means investors’ expectations for the pace of rate increases to slow after this month may be dashed, with rates potentially ending the year a half-point higher than most currently assume.

Write to Lisa Beilfuss at [email protected]