by Byron Gangnes
Since it keeps coming up, let me spend a few minutes explaining this.
Here is a basic summary of the approach used by the US Bureau of Labor Statistics (BLS):
For renters, shelter costs are straightforward: what they are paying in rent. For owner-occupants, the services provided by housing are estimated based on prevailing market rents (what is called "imputed rent: what you would pay to rent your own home.) That's because we want to know what housing services people are consuming in a given month, just as we measure the cost of other services consumed, like medical care, etc.
Buying the house itself is treated as an investment. It is a capital good that provides the housing services that we consume over time, as well as capital gains or losses when it is sold.
This is a consistent and economically sound approach that follows what most other countries do. It has been in place since 1983, with some tweaks since then, mostly to fine-tune how the sample of households is obtained for the underlying survey.
Here's a quote from the BLS methodology page (link to the page is below):
"The shelter service that a housing unit provides to its occupants is the relevant consumption item for the CPI. Most of the cost of shelter for renter-occupied housing is rent. For an owner-occupied unit, most of the cost of shelter is the implicit rent that owner occupants would have to pay if they were renting their homes...."
"Owned housing units themselves are not priced in the CPI Housing Survey. Like most other nations' economic statistics programs, the CPI program views owned housing units as capital (or investment) goods distinct from the shelter service they provide, and therefore not as consumption goods. Spending to purchase and improve houses and other housing units is treated as investment and not consumption in the CPI."
One issue that has gotten much attention recently is that the estimated rents, and therefore imputed rents, are based on current and a limited number of past (lagged) values of market rents. Note that data like Zillow's on rents is only for new asking rents. Some renters in each BLS survey are in the middle of leases, and so they do not change often. Effectively, it takes about 12 moths for market rent increases to fully filter through.
Of course, home prices increases themselves do eventually get reflected in the CPI, to the extent that they raise market rents.
Here is a link to the rather technical methodology page on the BLS web site. And a less technical overview of the issues from a Brookings Institution blog post.
Measuring Price Change in the CPI: Rent and Rental Equivalence, US Bureau of Labor Statistics, September 8, 2023.
How does the Consumer Price Index account for the cost of housing? David Wessel and Sophia Campbell, May 18, 2022.
Note: This post was also provided on LinkedIn.