Inflation has substantially cooled from its highs in 2022. However, it remains above the Federal Reserve’s preferred rate of 2 percent, in large part because housing inflation for both rented and owner-occupied housing is only slowly abating. Economists expect that housing inflation will continue to fall in 2023, further reducing overall inflation. This expectation has to do with the way prices of rental leases factor into the way inflation is measured.
The underlying argument goes something like this: Price changes for housing—rent and the cost of owning a house—are the single largest item in the U.S. Bureau of Labor Statistics’ (BLS) Consumer Price Index for All Urban Consumers (CPI-U).* The measures of both rent and the cost of owning a house in the CPI-U depend on the way the index for rental leases is constructed. This index encompasses more than just new rental leases; it takes time for the index to capture changing market conditions, reflected in new rental leases. Therefore, the rental price index, housing inflation, and overall inflation lag behind changes in prices for new tenants’ leases by about 12 months.
The prices of new rental leases have been falling for more than 12 months, but those declines are not yet fully reflected in the rental price index in the CPI-U.
Importantly, the prices of new rental leases have been falling for more than 12 months, but those declines are not yet fully reflected in the rental price index in the CPI-U. Current BLS inflation rates are higher than would be the case if the index only measured price changes for new leases. The flipside is that the measure for all rents—not just new ones—will eventually also fall and thus likely contribute to further declines in reported inflation.
This column explains why the rental price measure included in the overall price index lags behind measures for new leases, which have shown falling inflation for a while.
What is the real inflation rate?
Consumer Price Index for All Urban Consumers
The CPI-U is a widely watched measure of inflation in the United States. It captures the prices that consumers pay for a range of products, including new and used cars, prescription drugs, clothing, and services such as doctors’ visits and housing. Technically, the CPI-U is a weighted average of the prices for the typical goods and services that the average American household pays for in a given month. It can increase if either the price of a really important item, such as energy or housing, increases substantially or if the prices of many items go up.
This is what happened in late 2021 and early 2022 as the economy emerged from the pandemic. The prices of many items—from meat and eggs to cars, travel, and housing—dramatically increased as the economy reeled from pandemic-related supply chain bottlenecks, increasing consumer demand related to a resurgent labor market and changing consumer spending patterns away from services. Over time, things have settled down: Supply chain bottlenecks have eased, consumer spending has eased, and people have started to spend more money again on services such as health care and less on goods such as gasoline and electricity. This suggests that inflation has subsequently cooled from its highs in mid-2022. (see Figure 1)
CPI-U based on three-month averages
Economists typically average price changes over periods longer than one month since monthly inflation can be very volatile, and longer averages—such as three-month changes—better show the trends. Inflation started to surge in September 2021 and continued until it reached its peak in June 2022. Annualized inflation based on three-month averages has since declined from a high of 10 percent in June 2022 to 2 percent in the most recent three months, from March 2023 to May 2023. (see Figure 1)
Core inflation
Ignoring some of the most volatile price changes when measuring inflation can help provide a better view of the underlying trends. The BLS reports a CPI-U without energy and food—two of the most volatile parts of the economy—for that reason. This index is often referred to as “core CPI,” and its changes are subsequently called “core inflation.” The three-month average core inflation has declined from a high of 9.2 percent in June 2021 to 4.8 percent in the most recent three months. (see Figure 1)
Supercore inflation
Since used car prices and measures of prices for shelter—rents and the cost of owning a house—have proven to be unreliable indicators of what is actually happening with prices from month to month, economists also consider an even narrower index that leaves out energy, food, shelter, and used cars. This index, sometimes referred to as the “supercore index,” encompasses a variety of services, such as health care, hospitality, and education; and its changes are called “supercore inflation,” meant to convey the trend in the “real” underlying inflation. Notably, the three-month average supercore inflation has fallen from 6.8 percent in May 2022 to 2.4 percent in the most recent three months, just slightly above the Federal Reserve’s inflation target of 2 percent. (see Figure 1)
Changes in the cost of rent and owning a house either overstate or understate market conditions
Omitting shelter inflation—changes in prices to rented properties, owner-occupied housing, and lodging away from home, such as hotels and motels—from an overall inflation measure, alongside other key items, moves the CPI-U further and further away from how people actually experience inflation in their daily lives. But it may give policymakers at the Federal Reserve and in Congress a better sense of where prices are headed in the near future.
In particular, this approach raises eyebrows since prices for renting and owning a house or apartment, when combined, are the single largest item in the CPI-U. In fact, this measure accounts for almost a third of the total price index and is the largest expense for a family. Specifically, the BLS measures rental inflation, which also underlies the measure of owning a house, based on a carefully constructed sample of rental units, their current prices, changes in the condition of those units, and changes in utilities paid for those units. The BLS uses this rental index not only to depict rental prices but also to calculate the cost of owning a house, which is called “owners’ equivalent rent” (OER). This is simply meant to capture the implicit value of services that homeowners “consume” from their own homes, not the asset value of the house as an investment. OER is calculated using a subset of the same rental data as the CPI: Rent index, weighted by the price that homeowners think their home could be rented monthly, unfurnished and without utilities. This approach approximates the cost of living in a house—captured by the rent that a household otherwise would have had to pay—and it makes inflation less volatile since price swings in the housing market do not affect it. However, this also means that the rental index informs measures for both renting and living in one’s own house and, hence, accounts for almost a third of the overall CPI-U.
The BLS constructs the rental index to measure the lived financial experience of all renters. In contrast, leases for new tenants reflect the prices that landlords think their properties are currently worth. Many renters will continue to rent their same residence and may face smaller rent increases than new tenants if landlords value stability in their tenancy. Prices for new leases, therefore, often increase faster than average rents, particularly in an inflationary environment, as was the case in 2021 and 2022. Changes in two measures for new leases, for example, grew much faster in early 2022 than the rent index that is part of the CPI-U. Specifically, the Zillow Observed Rent Index (ZORI) equaled an annualized rate of 15 percent in the first quarter of 2022, while the marginal rent index reflected an annualized rate of 12 percent during that same time.
Meanwhile, the consumer price index for rents averaged only 5.5 percent, as it often takes some time for the level of leases for continuing tenants to catch up to the level of new leases—the current market prices. This also means that the average rental price index in the CPI understated rental market conditions during the run-up of rents in late 2021 and early 2022 and inversely understated the slowdown in rental inflation starting in late 2022.
See also
Declining new lease increases could contribute to falling average rents and slower inflation
Importantly, measures for new leases have shown declining inflation—smaller and smaller price increases—since September of 2021. These ever-smaller price increases, falling from lofty heights, have been below the changes in the CPI for rents, as measured by the BLS, since November 2021.
In short, prices are still rising but at ever slower rates. This is especially true for new leases, as they have increased less than average rents, contributing to an eventual and gradual fall of rental price inflation in the CPI index. In general, researchers at the BLS and the Federal Reserve conclude that the average rental index—the CPI for rents—lags behind measures for new leases by about 12 months. Given that price increases for new leases slowed down before those for average rents, increases in average rents will probably continue to fall. And given that the average rental index accounts for about one-third of total inflation, overall inflation could decline as well.**
Notably, inflation would have risen sooner and faster—and fallen earlier and faster—than reported if the CPI had included a measure for new rental leases, rather than the measure for average rents that the BLS actually uses.
Conclusion
Market dynamics in the rental market indicate lower inflationary dynamics. This suggests that the Federal Reserve can proceed cautiously in its design of monetary policy as long as other parts of the economy do not exhibit accelerating inflationary pressures again.
The Federal Reserve can proceed cautiously in its design of monetary policy as long as other parts of the economy do not exhibit accelerating inflationary pressures again.
*Authors’ note: The CPI-U is a widely cited and used inflation measure, but it is not the only one. The Federal Reserve also uses, and actually prefers, the price index for personal consumption expenditures (PCE) that is part of the Bureau of Economic Analysis’ national income and product accounts (NIPA). The arguments presented in this column only pertain to the CPI-U, not to the PCE price index.
**Authors’ note: This assumes that no other substantial part of the CPI-U, such as medical services or energy, exhibits an accelerating trend in prices, offsetting the decline in housing inflation.