Part of a central bank’s challenge in setting its policy rate lies in knowing which measure of inflation to follow. The consumer price index (CPI), the personal consumption expenditures (PCE) price index and other measures of inflation can each tell a different story.

While the headline CPI rate has finally dropped below 3%, falling to 2.9% in July, investors and policymakers are closely following the Federal Reserve’s preferred variable in setting interest rates, the PCE index, which settled in at 2.5% in July.

The core PCE index, which excludes the more volatile food and energy components, dropped to 2.6% on a year-over-year basis in July, below its 2.8% average since January. 

But both the CPI and the PCE index fail to fully capture inflation’s full picture. To be more specific, both the CPI, which the public watches, and the PCE index, which the Fed prefers, do not fully capture housing inflation, which has been slow to come down to the Fed’s 2% target.

To better gauge the impact of housing, we look to another measure of inflation, known as the harmonized index of consumer prices (HICP). 

It’s an index used in the European Union to track, or harmonize, pricing dynamics across countries.

Using the HICP, inflation in the United States stood at 1.7% in July, according to the Bureau of Labor Statistics. If the Fed used HICP in setting its policy rate, it would have most likely joined the parade of global central banks cutting rates at its July meeting rather than waiting till September, as it has signaled.

So what is the difference between the HICP and the better-known CPI and PCE index?

First, the harmonized index includes the U.S. rural population; the CPI, by contrast, focuses primarily on urban residents, who make up 87% of the American population.

Second, and more important, the CPI and PCE index exclude owner-occupied housing and instead create what is called the owners’ equivalent rent series, which tries to identify the cost of shelter service that an individual housing unit provides to its occupants. This approach attempts to measure the ongoing consumption of housing services in contrast with the change in the value, or price, of the home itself.

The fictive owners’ equivalent rent series attempts to estimate the imputed rent that owners would pay if they were renting their homes, excluding furnishings and utilities. This metric does not include things like mortgage interest, taxes, fees, home improvements or maintenance in the estimate of housing inflation.

The HICP approach, by contrast, measures the cost of owner-occupied housing by focusing on the cost of acquisition. That cost includes the price of new and existing homes bought by people for their own use, excluding investment properties, and includes the cost of repairs and improvements. Mortgage interest payments are considered financial transactions and are not included in the HICP estimate of inflation.

Essentially, the HICP measures actual housing costs, while the CPI measures what the Bureau of Labor Statistics thinks homeowners pay. In the end, the CPI is a good-faith estimate.

There is a risk to using owners’ equivalent rent. If the Fed holds rates too high for too long as it waits for disinflation to show up in the CPI data, it risks prematurely ending the business cycle based on a fictive metric of housing inflation. Housing inflation, along with services-based prices, is the primary cause of the difference in the Fed’s policy variable at 2.5% in July and the HICP at 1.7%.

This difference is a well-known issue within the community of central bankers and economists. While the Fed will not be using the HICP to officially make policy during this business cycle, it should consider harmonizing its policy framework with the other central banks after this cycle ends.

This is one reason why we have consistently pointed out the vast difference between what a near real-time housing metric such as the Cleveland Fed’s new tenant rent index shows and what appears in the CPI and the PCE index, which operate with a 12-to-18-month lag.

This lag has resulted in the Fed waiting to see the whites in the eyes of disinflation, an approach that if carried out too long risks becoming a policy error that tips the economy into an unnecessary recession. 

The different inflation metrics used by the Federal Reserve, the Bank of England and the European Central Bank are resulting in differing interest rate policies.

Rate cuts by the European Central Bank in June and the Bank of England in August are already causing a divergence in currency valuations. The U.S. dollar is up strongly against 15 of the 16 major trading currencies this year. And the divergence is a reason that roughly 30% of international currency flows are heading for U.S. markets and its higher rate of returns.