Does CPI Reflect True Inflation? Some on Wall Street Have Doubts.

“Owners’ equivalent rent” supplanted house prices in the 1980s in the construction of the consumer price index. The change had the effect of dampening the impact of soaring house prices. (Eric Thayer/Bloomberg)

After President Donald Trump’s sacking of the commissioner of the Bureau of Labor Statistics in the wake of a weak July employment report, investors will be parsing this coming week’s consumer-price-index release even more keenly.

The CPI is the most widely followed of all inflation measures, used to set many highly visible gauges such as cost-of-living increases for Social Security, even if the Federal Reserve uses a different index as its inflation guide. For investors, the CPI is key because it determines the inflation adjustment for Treasury inflation-protected securities, or TIPS.

“The $2.1 trillion market for TIPS is built on a foundation of trust in the construction of the CPI data, which is produced by the BLS. As such, the integrity of this data is at least as important as the employment data,” Michael Feroli, chief U.S. economist at J.P. Morgan, wrote in a research note this past week.

“In a scenario where TIPS are unable to provide a hedge against inflation because of the uncertainty around the CPI index published by BLS, its overall demand might start decreasing,” wrote Morgan Stanley fixed-income analysts Aryaman Singh and Matthew Hornbach in a client note.

In other words, if the market senses the CPI is being suppressed artificially, it will exact a penalty in the form of a lower price for the securities, which translates into a higher yield.

The import extends beyond these inflation-indexed securities. “Quite simply, bonds and TIPS are joined at the hip,” Rob Arnott, chairman and founder of Research Affiliates and a longtime analyst of inflation-indexed securities, said in an interview.

The real returns of both types of securities are affected identically, which in turn affects the valuations of equities.

Manipulation of government-issued economic statistics would hardly be unprecedented. Chinese government data has a long history of hitting official targets nearly as accurately as General Electric used to top earnings-per-share estimates by a penny in the Jack Welch era.

Argentina routinely manipulated its inflation numbers in the past decade as prices soared. And Turkey’s official inflation numbers, reflecting an 80% jump in prices, were less than half what private economists published earlier this decade. Turkish President Recep Tayyip Erdogan responded by firing the central bank officials who opposed his bass-ackward insistence that high interest rates caused the inflation.

What’s more nettlesome to Arnott is how the calculation of the U.S. inflation numbers keeps getting changed, invariably to produce a lower number by diminishing the most bothersome cost pressures.

Perhaps the most notorious example was how the “core” inflation concept was born during the initial surge of the Great Inflation of the 1970s. Following OPEC’s quadrupling of oil prices and a sharp rise in food prices, so-called core inflation excluded food and energy prices, which were deemed to be outside of the influence of monetary policy. That the surge in commodity prices followed the severing of the dollar’s value to gold in August 1971 was dismissed by policymakers.

But Arnott points to changes in how the overall CPI is calculated, not the core version. In the early 1980s, house prices were supplanted in the index by the hoary notion of “owners’ equivalent rent,” a construct that asked homeowners to estimate how much they would pay to rent their own houses. That has had the effect of dampening the inflation impact of soaring house prices, a painful fact of life for people trying to buy a home.

Arnott credits former Treasury Secretary Lawrence Summers (with whom he rarely agrees) for pointing out how the BLS rule changes from four decades ago masked the Covid-era inflation. Under the current rules, the year-over-year CPI change hit 9.1% in 2022 as home buyers bid up prices with 3% mortgages. That was the highest since the 14.7% peak in the CPI.

But, Arnott observes, under the old method, the CPI would have registered a 19.5% surge, even worse than the peak inflation of the early 1980s, which spurred the Fed under Paul Volcker to lift interest rates to 20%.

Economists debate the true inflation impact of house prices. How much reflects the cost of living and how much is an asset price, like stocks? (Let’s leave the question of asset inflation for another day.) But the BLS’ calculation that healthcare costs have declined over the past five years utterly beggars credulity, Arnott says. The agency comes to that conclusion from an estimation of insurers’ profits. “That’s insane,” he declares.

For investors, the prospect that the CPI may not reflect true inflation should diminish the price of TIPS, Macquarie strategists Thierry Wizman and Gareth Berry write in a client note. In turn, manipulation of the CPI could steepen the U.S. Treasury yield curve. That is, longer-term bond yields’ premium over short- and intermediate yields could increase to compensate for higher future inflation, which could be understated.

If that were come to pass, it would thwart the Trump administration’s aim to lower the 10-year Treasury yield, a benchmark for many private borrowing costs, notably home mortgage rates. If so, it would backfire in a similar manner as pressure on the Fed to lower its short-term policy rate, which has been followed by higher bond yields.

The suppression of reported inflation would mean inflation-indexed investments such as TIPS would underperform, the Macquarie strategists write. A better hedge would be gold, which should benefit if the CPI is manipulated. In that case, you would want real assets whose compensation doesn’t rely on “official” inflation data, they conclude.

Write to Randall W. Forsyth at randall.forsyth@barrons.com