This article discusses how owner-occupied housing is factored into the consumer price index (CPI) and analyzes the role CPI plays in the real estate industry.
The basics of CPI
Vision is always 20/20 in hindsight. Had we known at the beginning of the millennium what we know now, millions of people wouldn’t have been hurt by the sudden explosion and collapse of home prices. Had we known of the massive damage that would be inflicted by the swelling of asset prices, preventative measures would have been put in place to deliberately avoid it. But we don’t possess the luxury of omnipotence. Instead, we must rely on careful analysis and forecasting to divine the economic contours of the coming years.
However, forecasting is entirely dependent on the data used for analysis and the analytical bias, whether optimistic or pessimistic, held by the forecaster. Erroneous or incomplete data will inevitably lead to erroneous conclusions, and innate biases will interfere with the forecaster’s perception of reality. Which begs the question: could any of us have predicted and avoided the ultimate disaster of the Millennium Housing Boom had we been tracking the movement of price inflation?
This question is proposed by Erwin Diewert and Alice Nakamura in their research paper for the Federal Reserve Bank of Philadelphia titled, “Accounting for Housing in a CPI.” They argue the rental equivalence approach – comparable rental value – now in use to account for owner-occupied housing services (OOH) in the consumer price index (CPI) fails to accurately measure inflation in OOH, and thus a new approach is needed to recognize and avoid future market anomalies.
The periodic percentage change in the CPI is the most universally recognized measure of domestic inflation and is used to measure price movements of goods and services associated with the consumer’s cost of living, not the ownership of capital assets. At its core, the CPI tracks the change in the price paid by consumers for goods and services based on the price paid for the same goods and services in the prior year. As the CPI is concerned only with goods and services, investments and assets are intentionally omitted from the calculation. No asset inflation index (AII) exists, though such an index would be helpful to brokers for pricing property and stocks.
The central difficulty in calculating OOH costs in the CPI is that OOH functions on two levels: as a durable good, being the home which provides constant shelter for the occupying owner, and simultaneously as a capital asset, which does not belong in the theoretical framework of CPI. Thus, when OOH is factored into CPI as a consumer cost, OOH’s dual nature poses a uniquely difficult and highly contentious challenge to get it right.
The presence of OOH in CPI is particularly important as the cost of housing constitutes a considerable portion of a household budget (typically more than 30% for recent buyers of homes), and remains so for a period of time longer than most other consumer goods.
An evolving history
The calculation of OOH in CPI is as controversial as it is important. The Bureau of Labor Statistics (BLS) and Bureau of Economic Analysis (BEA) have made continual changes to their methodology for estimating OOH in CPI, and disagreement still flares in the business, labor and academic communities regarding its computation.
From the early 1950s to the 1980s, shelter costs in CPI were calculated using the asset price approach. The asset price approach weighs the costs associated with purchasing and maintaining a house as a physical asset, including:
- home purchase price;
- mortgage interest costs;
- property taxes;
- homeowners’ insurance costs, and
- maintenance and repairs.
However, the asset price approach contained a fundamental flaw. As the name implies, the measure was largely reliant on purchase price and mortgage rates, asset elements which are not to be considered when calculating OOH in the CPI.
In the early 1980s, The BLS considered but did not adopt a user cost approach to quantify the cost of shelter provided by the home owned by the occupant. Under the simplified user cost approach, OOH is determined by adding the cost of purchasing a home at the beginning of the year, the cost of inhabiting it for a year (including the cost of mortgage interest, property taxes, property insurance, maintenance, return on equity and value appreciation), then selling the property at the end of the year. In essence, the user cost approach accounts for the costs the property owner actually incurs by providing his own housing.
However, the acquisition of this data proved too onerous for national adoption by the BLS. At the time of its proposed adoption, home purchase prices and mortgage interest rates were only available on houses with FHA-insured mortgages, which were becoming a greater rarity. The user cost approach also suffered from the same asset specific indicators (price, mortgage payment, return on equity) as the asset price approach.
The computation of OOH in the CPI is equally mercurial on the global stage. Canada and Iceland both use modified user cost approaches, and many countries simple omit the cost of shelter completely in their factoring of the CPI.
Rental equivalency approach
Beginning in 1983 and continuing through the present day, OOH shelter components for the CPI are factored using the rental equivalency approach. Under the rental equivalency approach, which has been revised multiple times since its inception, the BLS computes OOH in the CPI based on implicit rent, the open market rent for a similar dwelling during the same period of time. To determine this, sampled households are given the Consumer Expenditure Survey which asks the following question:
If someone were to rent your home today, how much do you think it would rent for monthly, unfurnished and without utilities?
To divine the change in housing prices, the BLS examines rental units within a sample area consisting of three to four city blocks. The rental unit data is used to calculate the rental costs within the sample area and the implicit rent an owner-occupier would receive by renting out their property located within the sample area.
Similar to the asset and user cost approach, the rental equivalency approach indirectly accounts for the cost of mortgage interest, property taxes, property insurance and maintenance and repairs since all these factors are included in rent received by a landlord as the rental income pays for these operating expenses (and provides a return on capital).
As is commonly known in the real estate and appraisal industry, there are three fundamentals used to determine the value of a property:
- rents (what the property would generate if rented out on the open market under the rental income approach);
- replacement costs (the cost of land, labor and materials to completely rebuild the property from scratch); and
- comparable properties (the cost of similar properties within the immediate vicinity of the property which share fundamental characteristics of the subject property).
The rental equivalency approach is consistent with these three fundamentals. Agents and brokers employ the comparable market analysis (CMA) to set the price for rent in an area. [See first tuesday Form 318]
The rental equivalency approach is the most common approach used globally, employed by 13 of the 30 nations in the Organization for Economic Cooperation and Development (OEDC).
Editor’s note: It is likely that if more owner-occupiers were aware of the rental equivalency of the property they own, more would walk away once they discover they can rent the property next door for a fraction of the cost of owning the property they currently inhabit. This is especially true in places like California where a larger percentage of the population overextended themselves to afford housing and are now underwater.
A challenge from the opportunity cost approach
The rental equivalency approach is not without its detractors. The authors of the “Accounting for Housing in a CPI” at the Federal Reserve Bank of Philadelphia claim the rental equivalency approach did not accurately reflect the massive growth of the housing bubble which began in 2002. The observed rise in housing prices between 2002 and 2005 was not reflected by an increase in the implicit rent under the rental equivalency approach for those properties.
Editor’s note – One should be reminded it is not the purpose of the CPI to reflect asset bubbles, no matter how prevalent they may be. It is concerned only with the costs of living, such as providing shelter.
The authors of the Federal Reserve paper conclude that the rental equivalence and user cost approaches originate from the same theoretical framework. Thus, in theory, they should move in tandem; the fluctuations of rents should mirror the fluctuations in user costs for comparable dwellings. However, they don’t – rents are the basis for setting value/pricing; pricing is not the basis for setting rent in any property which is equivalent shelter.
The authors cite multiple reasons why user costs and rental rates have been shown to diverge:
- the stickiness of rental rates encourage many tenants to renew their leases to retain advantageous terms, preferable to renting a different unit;
- owners and renters encounter different types of uncertainty regarding long-term housing and operating expenses;
- the lack of high-tiered luxury homes in the rental market; and
- tax regulations treat owner-occupiers differently from landlords.
Editor’s note – This reasoning can be easily contested. Landlords demand market rates when they rent out a unit. A landlord’s obligation to maintain a unit to get maximum market rents from a new tenant is similar to the expense incurred by an owner to maintain the property he occupies in a habitable condition. The risks and uncertainties faced by owners and renters are largely the same – they both are conscious of value, interest rates and operating expenses whether they own or rent. Also, there in an obvious reason why there is a lack of high-tiered homes in the rental market: there is an equal lack of demand. As for tax regulations, taxes are unavoidable and must be paid whether you are an owner-occupier or a landlord, and both write-off interest and taxes (which isn’t the case in most countries).
The authors have formulated a new approach called the owner-occupied housing opportunity cost approach (OOHOC) and propose it should be used in place of the real equivalency approach. Under the OOHOC approach, the cost of OOH is determined as:
- the rental opportunity cost (the cost of renting an equivalent dwelling, generally called implicit rent); and
- the financial opportunity cost (the rate of return the owner-occupier could have received by intelligently investing funds but foregoes by owning a home, with mathematical adjustments made).
Editor’s note – An opportunity cost is concerned with financial assets and net worth, not the cost of items consumed such as shelter, and it should not be considered as having an impact on the CPI.
The rental opportunity cost is the comparable rental rate for the unit and functions the same as the rental equivalent approach (comparable rents). What distinguishes the OOHOC approach is the inclusion of the financial opportunity cost, the rate of return an owner could have received by smartly investing the funds that are instead tied up in the property. However, the proposed financial opportunity cost calculation contains various mathematical modifications to distinguish it from the opportunity cost approach.
For example, owner-occupied homes exhibit less physical deterioration than rented units since they tend to be better maintained by the occupiers. Thus, the physical age factor is removed from the financial opportunity cost calculation. Instead, the number of years the owner intends to stay in the home and continue paying on the mortgage is considered in the financial opportunity cost calculation.
The calculation considers whether the owner-occupier:
- owns the property free and clear;
- still owes on their mortgage though has a positive equity position in the property; or
- has zero home equity.
Editor’s note – None of these factors have any relation whatsoever to price or value and should not be included in the CPI calculation or an asset inflation index. For many property owners, the amount of the mortgage has become a money illusion and is thus employed by them to artificially set their property values.
Owners in a negative equity position, owing more than the home is worth, are not currently factored into the financial opportunity cost. However, if we are to accept the logical premise of the financial opportunity cost, negative equity must be considered since the Millennium Boom left many California homeowners upside down and underwater on their mortgages. Nearly 90% of homeowners with negative equity continue to pay and will eventually payoff the excessive debt (in lieu of other purchases to raise their standard of living and contribute to a recovery).
A rudimentary problem still exists with the OOHOC model: it computes a component of homeownership as an asset, which is inappropriate when considering CPI. The authors make the attempt to coerce the CPI into being an indicator of asset inflation, which by definition it simply is not. The authors would be well served to focus on this simple question: what’s the value of the right to use a property? It’s not the ownership status of a property that is important when considering the CPI, it’s the possession of a property, be it through fee ownership or leasehold. Monthly rent, the price for one month’s consumption by the occupant of shelter, is the most accurate reflection of the price paid for possessing a property, where as the financial opportunity contained in the OOHOC model has nothing to do with possession and everything to do with asset value.
Who uses CPI in the real estate industry?
The CPI plays an important role in price analysis for the real estate industry as it underlies many aspects of a sale, lease or loan transaction.
Landlords frequently rely on CPI to account for rental inflation in their leases. [See first tuesday Form 552]
In the lease agreement, the CPI annual adjustment for rents limits how far the rental rate can increase from the initial base year rent. Historically, rental rates move in tandem with CPI, though during any short-term period this isn’t necessarily the case. Rents are subject to booms and busts just like property prices. However, they are more stable and tend to cycle upwards less than property values during booms and decline less during bust years. Overtime, this movement reflects the average annual increase in the rate of inflation – as expressed in the CPI. [See first tuesday Form 552 §3.4]
CPI also affects lenders’ judgment. Lenders use CPI when they calculate the annual interest they need on a fixed rate loan, no differently than the calculation of interest expected on 10 or 30 year treasuries.
Adjustable rate mortgages (ARMs) do not reflect inflation and thus do not use the CPI, but change in accordance with one of multiple short-term cost of money indices, such as the LIBOR or COFI. However, in a lending situation, the CPI functions less as a literal inflation rate, but as the lender’s anticipated prediction of inflationary trends. Lenders want whichever cost of money index they select to consistently stay multiple percentage points – a profit margin – above their projected CPI to ensure they maintain a constant real (after inflation) return.
Asset price: the necessity of a new index
The CPI is subjective and messy, and the way it is calculated is constantly revised, critiqued and criticized. Many CPI pessimists accuse the BLS of changing their calculation procedure purely to disguise inflationary trends to keep entitlement programs and interest rates in check since government expenditures are affected by inflation as are California property taxes.
The calculation of OOH in CPI is not an exact science. However, CPI is not intended to be exact, rather, it is a general measure of inflation for expenditures associated with the cost of living. It is also a measure of consumers’ domestic purchasing power for the dollars they earn or hold. Though there has been a rising sentiment that the BLS should modify the way OOH is calculated in CPI, adopting a modification which treats any component of OOH as an asset would be fundamentally at odds with the stated purpose of the CPI. The creators of the proposed OOHOC approach keep running into the same conceptual brick wall – you can’t incorporate asset values into the CPI, and you can’t force the CPI to analyze that which it was never intended to analyze, asset inflation.
It’s impractical to twist the arm of the CPI in an attempt to force it to monitor something beyond its consumer boundaries. A more obvious solution would be the creation of an asset price index (AII) with the sole purpose of monitoring real estate and stock market investments for signs of excessive price movements. Though policy makers have the complete ability to prevent bubbles from forming in asset prices, what they lack is an accurate method of monitoring whether an aberrant and potentially damaging inflationary or deflationary condition exists so they can take those steps to mitigate it.
With a much needed and long overdue asset price index in place, regulatory agencies such as the Federal Reserve could determine which governmental actions are needed to limit inflationary or deflationary pressure to allow for stable long-term growth of the economy.