Modernizing the Consumer Price Index for the 21st Century (2022) / Chapter Skim
Currently Skimming:
4 Modernizing Difficult-to-Measure Expenditure Categories: Housing/Shelter
4 Modernizing Difficult-to-Measure Expenditure Categories: Housing/Shelter
Pages 85-104
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 85... ...
and medical care (close to 9 %) .1 Although shelter remains a CPI category with inherently difficult conceptual and measurement issues, new source data have become available that may create opportunities to improve price measurement for housing services.
|
From page 86... ...
Rather, their housing-related cost of living is the amount they would have had to spend in order to consume the same amount of housing services provided by their owner-occupied home.4 Since the rent paid by renter households is an observable stream of expenditures used to consume housing services, one common approach is to impute the price of housing services for owner-occupied housing using rent data, called the "rental 2 This emphasis is not meant to imply that there are not also important issues having to do with measuring price changes faced by renters. For example, several researchers have questioned the CPI method for sampling rental units -- particularly the mix of properties that have and have not changed hands.
|
From page 87... ...
4.2. RENTAL EQUIVALENCE APPROACH TO ESTIMATING PRICE CHANGE FOR OWNER-OCCUPIED HOUSING The basic premise of a rental equivalence approach is that the price of owner-occupied housing services can be imputed by estimating the amount that would be required for a household to rent the same home.
|
From page 88... ...
, it appears that single-family rental units are undersampled in the CPI Housing survey.9 This undersampling might reduce the accuracy of the imputation of rent changes to the owner-occupied stock since many such units are single-family dwellings. Recent research has found that rents of single-family and multifamily units within the same neighborhood did not rise at the same rate from 2013 to 2016 (Adams and Verbrugge, 2021)
|
From page 89... ...
Key housing characteristics that are missing from the ACS include the floor space of the structure and land area. Consequently, block groups with larger homes on more land may not be given enough weight in the CPI if the available characteristics un derestimate actual structure size and land area.12 One big-picture drawback of the rental equivalence approach is that if owner-occupied and rental markets are segmented, it can be difficult to find rental units that are reasonably comparable to owner-occupied homes in the same block group.
|
From page 90... ...
However, if consistency is the goal, pricing other durable goods on a flow of services basis is another option.14 4.3.2. User Cost Approach Another option for measuring price change in owner-occupied hous ing is the user cost approach.
|
From page 91... ...
argued that the best approach to valuing the services of owner-occupied housing is to take the maximum of its rental value and its user cost value, using long-run expected inflation rates for expected capital gains in the user cost formula. They refer to this method as the "opportunity cost" approach because it attempts to quantify the opportunity cost of occupying a home instead of renting it out.
|
From page 92... ...
The first is theoretical. Although the expenditures used in the payments approach are components of the user cost, the pay ments approach does not include all of the elements of the user cost, such as the opportunity cost of owners' equity, depreciation, and expected capital gains.
|
From page 93... ...
. 18 The case against using the payments approach is detailed more completely in the draft chapter by Diewert and Shimizu for the International Monetary Fund CPI Manual www.imf.
|
From page 94... ...
The rental equivalence approach is consistent with the cost-of-living index objective of the CPI, and the change over time is based on observed price changes. The user cost approach is also conceptually consistent with a cost-of-living index objective but is a model-based approach that has some practical operational constraints.19 The acquisitions approach and the payments approach are not fully consistent with a cost-of-living index objective.
|
From page 95... ...
They noted, however, that the CPI methodology may be appropriate for an index with a purpose of measuring cost of living, since rental market contracts and frictions do, in fact, reduce the influence of market conditions on the rent that households actually pay. Beyond the potential of these data sources to improve the CPI's abil ity to reflect rapid changes in rent growth, resources saved from collecting multifamily data in the CPI Housing Survey could be used to expand the survey to include more single-family units in owner-occupied neighbor hoods.
|
From page 96... ...
On their own, property tax records could not be easily used for the purpose of estimating expenditure shares of owner-occupied households because it is difficult to distinguish between rental units and owner-occupied units in these data. However, the property tax records could be merged with rent data from the CPI Housing Survey or the ACS to impute rental expenditure to the owner-occupied stock.
|
From page 97... ...
It would be useful to publish shelter indexes for a larger set of locations, even if price changes for other goods and services could not be computed for these areas.22 Updating the geographic sample of the CPI every 10 years may not be sufficient to capture important changes in the geography of housing, for example the reported migrations spurred by the pandemic. There could be advantages to maintaining a larger and more diverse sample of housing units across cities even if other CPI items are not priced in those areas.
|
From page 98... ...
In these cases, a user cost approach might be helpful to improve estimates of the price of housing services. Further research comparing user cost estimates to rental equiva lence estimates would be valuable in helping BLS learn about the types of housing units and/or markets for which each approach would be preferable.
|
From page 99... ...
Research conducted on data at the micro level would be valuable for testing where the rental equivalence method is performing well and where it is having troubles, such as for pric ing housing services associated with higher-end properties. Recommendation 4.6: As part of its research program, BLS should compare rental equivalence estimates to user cost estimates for indi vidual properties, and also explore the opportunity cost approach.
|
From page 100... ...
For the purpose of creating alternative measures of housing-related inflation, different methodolo gies could be used depending on the purpose of the index. The top priority should continue to be improving data sources and methods to improve the rental equivalence approach that is suitable for the flagship CPI-U.
|
From page 101... ...
The tenant units would be resampled at six-month intervals to capture rent changes. Originally, the intention may have been to resample the owner units less frequently (maybe once every couple of years)
|
From page 102... ...
While it was thought possible that the difference reflected a real difference in the inflation rates of the types of tenant units that were most similar to owner-occupied units, there was also a concern that there may have been a flaw in the matching process, which was considered to be a bit of a black box. In the imputation algorithm, the six-month rent changes for OER were calculated using a Carli formula -- that is, the imputed owner's equivalent A00858 -- Consumer Price Index REV.indd 102 8/15/22 11:54 AM
|
From page 103... ...
St. Louis Federal Reserve.
|
From page 104... ...
While the panel does not know the full details of the motivation behind the shift, it seems that, at least in part, the new approach was driven by an understand ing that it was problematic to attempt to closely match owner units to similar tenant units -- especially in neighborhoods that were predominantly owner occupants -- and that the most practical approach was to attempt to reweight the tenant units to more closely resemble the population of owner units. A00858 -- Consumer Price Index REV.indd 104 8/15/22 11:54 AM
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.