Bengali, Leila. "Comparing Measures of Housing Inflation, 2022-29," Economic Letter (Federal Reserve Bank of San Francisco) (October 17, 2022). https://fraser.stlouisfed.org/title/4960/item/668468, accessed on January 20, 2025.

Title: Comparing Measures of Housing Inflation, 2022-29

Author: Bengali, Leila
Date: October 17, 2022
Page 1
image-container-0 FRBSF Economic Letter 2022-29 | October 17, 2022 Comparing Measures of Housing Inflation Research from the Federal Reserve Bank of San Francisco Leila Bengali Measuring the price of shelter for homeowners is difficult, even when housing markets are stable. A new measure of shelter price inflation uses mortgage, tax, and insurance payments, rather than the implied rental value of homes used in the consumer price index (CPI). The payments method suggests year-over-year shelter price inflation rose 4.3% nationally in July, compared with the CPI’s 5.8% estimate. Conditions in rental markets likely explain this difference. Comparing the varying results nationally and across regions highlights the challenge of accurately measuring the shelter inflation that homeowners face. Changes in housing costs for homeowners are accounted for differently in various inflation measures. For example, because the consumer price index (CPI) focuses on consumption, it considers housing as a service that people use, as opposed to an investment good, and only incorporates the price of that service into the index. Accurately measuring the prices of “shelter services” for homeowners is important for a few reasons. Shelter receives the largest weight among components of the CPI, so an accurate measure of shelter prices homeowners face means a more accurate inflation estimate. Moreover, shelter prices may be more directly responsive to monetary policy than prices of other goods and services because mortgage rates can adjust quickly in response to changes in policy interest rates set by the Federal Reserve (Kliesen 2022). Unfortunately, the price of shelter services for homeowners is hard to measure because the price cannot simply be read off a price tag. The price must be estimated, and there is no single best way—in fact, the method the CPI uses has changed over time. The current measure uses market rents, which assumes renters and owners face similar price dynamics. In addition, rents often change on an annual schedule, which can result in the well-documented lag between the CPI measure and price indexes for houses and rental units (for example, see Lansing et al. 2022). In this Economic Letter, I create a measure of shelter price inflation for homeowners based on monthly payments for mortgage, tax, and insurance that is meant to complement and address limitations in the CPI measure. My method finds national homeowners’ shelter prices rose 4.3% year-over-year in July 2022, compared with 5.8% in the CPI measure. I explore where and why these differences may occur, with conditions in rental markets playing a key role. Measuring the price of shelter for homeowners Estimates of shelter prices for homeowners aim to capture the price of the services provided by an owner’s home. The CPI uses a rental equivalence measure, called owners’ equivalent rent of primary residence. The measure is based on the “implied rent” that owners indirectly pay to live in their homes. Implied rent
image-container-1 FRBSF Economic Letter 2022-29 | October 17, 2022 2 cannot be observed, so the CPI uses an estimate. This estimate starts with actual rent data for occupied units, then uses econometric models to adjust the rents to account for differing characteristics between renter- and owner-occupied properties, such as the size and number of bedrooms. House prices and an owner’s monthly expenditures, such as mortgage payments and property taxes, are not included because these expenditures are the cost of making an investment rather than the price of consuming a shelter service, and the CPI is intended to measure prices of consumption goods and services, not investments. For this Letter, I turn to user cost measures as another way to estimate shelter prices for owners. These calculate the costs a homeowner incurs when buying and living in the home and can be justified using the same theory that validates rental equivalence methods (Diewert and Nakamura 2009 and Garner and Verbrugge 2009). I create a proxy for user costs with data on several of a homeowner’s major monthly shelter expenses: mortgage principal and interest payments, property taxes, and insurance payments. I calculate the average monthly payment in each U.S. county or metro area using a large random sample of mortgages from Black Knight McDash, a company that compiles data from residential mortgage servicers. An advantage of this method is that the payments measure uses actual homeowner costs, rather than estimated implicit rent. This sidesteps issues that could arise in the CPI’s rent-based measure resulting from differing shelter price dynamics for renter- and owner-occupied units and from the lag caused by the tendency for rents to change only annually. While the CPI is available for only 23 metro areas, the payments measure covers nearly all counties and metro areas. This is an advantage because, as a non- traded service, shelter prices show more geographic variation than prices in other CPI categories. The payments measure faces a few disadvantages. For one, it does not provide information about homeowners without mortgages. Also, because the data don’t include home characteristics, the payments measure cannot be adjusted to create a constant-quality index, which is a key element of the CPI measure. Changes in the payments measure reflect both changes in mortgage terms, such as refinancing, and changes in the sample composition. For example, if homes with newly originated mortgages become larger, are improved over time, and cost more as a result, inflation calculated using an unadjusted measure may be higher than with a constant-quality measure. Finally, to be a perfect proxy for user costs, the payments measure would need additional terms that are hard to quantify, such as depreciation, opportunity costs of investments not made, and expected capital gains on the home. In fact, as addressed in Bolhuis, Cramer, and Summers 2022, the BLS switched away from a cost-based measure in 1983 in order to address some of these measurement issues. Weighing the advantages and challenges of the payments method suggests it is a useful complement to, rather than replacement for, the CPI and can add to the information available for assessing homeowners’ shelter price inflation. Homeowners’ shelter price inflation by two measures Rental equivalence measures, such as what is used in the CPI, are theoretically equivalent to user cost measures. Though the payments measure is an imperfect proxy for user costs, its correlation with the CPI
image-container-2
image-container-3
image-container-4
image-container-5
About
Diversity is critical to the Federal Reserve, and we are firmly committed to fostering a diverse and inclusive culture throughout the Federal Reserve System. Collections within FRASER contain historical language, content, and descriptions that reflect the time period within which they were created and the views of their creators. Certain collections contain objectionable content—for example, discriminatory or biased language used to refer to racial, ethnic, and cultural groups. These viewpoints and attitudes are inconsistent with our values, but the original descriptions are retained to ensure that they are not erased from the historical record.