Who do YOU Believe?

Tracking Home Prices - Different Measures; Different Approaches,
by NAR Research Staff

Since all real estate is local, interpreting changes in home prices at the national or metropolitan level can be difficult. This can be problematic for policy makers and market participants like home buyers, sellers, and REALTORS®. The first step is understanding what the various measures of house prices are, what those measures actually track, and how this information can be interpreted and used. Below is a summary describing four most closely watched home price indexes and discussing the major differences between them.

There are four major home price indexes that market participants follow:
· the National Association of REALTORS® Median Sales Price series
· the Case-Shiller Index group
· the Federal Housing Finance Agency (formerly OFHEO) Index
· the Freddie Mac Conventional Mortgage Home Price index

Of the series listed above, only NAR's is a pure median measure. The NAR Median Sales Price (which is not an index) tracks the sales price of the typical home where half of the homes sold at a higher and half sold at a lower price in a particular month.

Indexes track home prices with a level of abstraction since there are several ways that information about home prices can be aggregated to develop an index. Consequently, changes in the prices of homes in a particular neighborhood may differ from an index based on home sales from a larger or slightly different geographic area.

Each of the other measures is an index. Like all indexes, these indexes track house prices with a level of abstraction. Percent changes in the index figure mirror percent changes in the prices of houses; therefore, a single index figure may not be very meaningful. The weighted repeat sales technique-employed by all of the other major players attempts to address an issue arising in pure sales medians, namely, that the houses that sell one month are not the same houses that sell the following month. If these houses differ in quality, then a change in the median sales price of houses reflects the change in the quality of the home sold as well as any appreciation in value that would have accrued to a constant-quality house.

Researchers track and measure repeat sales to get past this limitation. For example, if a house sold in 1980 for $100,000, and it sold again in 2007 for $400,000, it would register as a repeat sale. Researchers would use its 300% appreciation over 27 years with information from other repeat sales to determine how much this same (and therefore "constant quality") home had appreciated in the most recent time period.
Unlike a median sales price, the repeat-sales methodology is very complex and requires an extremely large database of home sales transactions that limits its application to the nation or to large metropolitan areas.

The National Association of REALTORS® Median Sales Price (NAR's median sales price) data is a series of dollar figures released monthly at both the national and Census regional levels. In addition, NAR releases quarterly median home price data for more than 150 metropolitan areas. NAR reports the observed median sales prices based on closed sales transactions gathered from Multiple Listing Services (MLSs). NAR's median sales price data on existing homes are available at the Association's web site at http://www.realtor.org/research. Click on "home sales statistics" in the left hand navigation.

Case-Shiller
The Case-Shiller Index group consists of 20 regional (metro area) indexes and two composite indexes (a 10-region and a 20-region index). These indexes are reported monthly as 3-month moving averages and cover a period ending two months prior. This group also consists of a broader national index that is released quarterly.
Case-Shiller uses a weighted repeat sales methodology to measure price change. This type of index examines price changes for the same home based on the two most recent sales transactions. While this methodology helps to control for several factors that influence price, it requires the exclusion of many data points. For instance, Case-Shiller indexes exclude data from foreclosures as well as home sales transactions in the following states: Maine, Indiana, Wisconsin, North Dakota, South Dakota, South Carolina, West Virginia, Alabama, Mississippi, Idaho, Montana, Wyoming, and Alaska. Additionally, Case-Shiller gathers information on transactions from local government sources and therefore, information about sales transactions is subject to delay. Furthermore, the index is value-weighted: that means that high-priced homes have a greater impact on the index than do low-priced homes. Sales financed with conventional as well as jumbo and subprime loans are included in the index.
For more information about the Case-Shiller Index, visit www.homeprice.standardandpoors.com.

continued in future posts