One of the most influential and widely used statistics to measure price changes in the real estate market is the S&P/Case Shiller Home Price Index.  It seems that almost every news report or article published uses this index as the foundation for the news story.  Some of the news stories report the facts of the index and leave the reader to interpret the data, while others provide commentary and conclusions that can be anything from valid, to misleading, or even grossly erroneous because they simply do not understand how the index is constructed. Given the ubiquity of the index, it is important to understand what the index is designed to tell us as well as what it conceals.

 How it’s Calculated

The Case-Shiller Home Price Index is widely recognized as the most accurate measure of home price change in a market because it tracks repeat sales of the same home over time and attempts to hold quality constant between these sales, however it is far from being the perfect measure that many hold it to be.  While there are many intricate assumptions within the index that can be debated both for and against, this article will provide an overview of how the index is constructed and illustrate its strengths and weaknesses.

The Index is calculated by collecting data on recent home sales that have a history of being sold previously.  If the same home has transacted in the past, it matches the price previously paid to the recent price to calculate the change in price.  It is important to know that not all repeat sales available are actually included in the calculation of the index.  For example, any homes with more than 1 transaction within a 6 month time frame are excluded from the index, any transaction that appears to be a pricing anomaly meaning that it does not fit the overarching statistical distribution of sales very well is excluded, and non arms length transactions are generally excluded as well.  Homes with long periods of time between sales are still included but are counted at a lower weight based on the assumption that the longer the home goes between sales the more likely it is that the home has changed in quality due to being remodeled or lack of maintenance.

The data is then aggregated up to a Metropolitan Area and reported on a monthly basis.  However, this monthly reporting is somewhat misleading in its implied precision.  The data being reported monthly is actually a 3 month rolling window of data.  For example, the index calculation reported in April actually includes all transactions from January, February, and March.  This is done to ensure a large enough sample size is available to report, but it also has the effect of smoothing out the trend line for the index.  Since each new month includes 2 moths of data that is also included in the previous months score, it puts a constraint on the amount the index can change from month to month.  The new months data will only have approximately 1/3 the impact on the total score.  This is an important assumption that is either not well know or simply ignored in many news articles published citing the index.

 What does it Reveal?

Being a repeat sales index, the Case-Shiller Index methodology provides a very accurate measure of price changes generalized to an entire market under the assumption of holding quality constant over time.  The index automatically adjusts for different physical characteristics of the sample from month to month thus allowing easy comparison of data over time.  It is indeed one of the best measures of the overall value of the housing stock in a particular market at a given time.

 What does it Conceal?

The focus on accuracy comes with a cost.  The methods used to construct the index also work to make it less sensitive to changes in the market.  It is not that the index will not pick up the price changes (if that were the case it would not be accurate), but rather it is very slow to show or react to price changes in the market.  Being a repeat sales index, it does not include pricing for new homes because they are first time transactions and have no data to compare it to.  The problem lies in that new home pricing is generally the quickest to react to market pricing conditions both on the way up as well as on the way down.  The omission of this type of data reduces the sensitivity of the index to detect changes in the direction of price appreciation.  New home prices began falling while the Case-Shiller Index was still rising, and the reverse will be true as well.  This builds in a lag time for new home pricing to set the stage for the resale market.  This lag is then compounded by the 3 month moving average methodology to calculate the index score.  Add it all up and the reality is the Case-Shiller Index is a lagging indicator and could be as much as 6 months in the rear view mirror so to speak.

In short, the Index is a very good measure of price changes over a given timeframe for a large geographic region holding the quality of the homes constant but is relatively weak in its ability to quickly detect and respond to changes in the market from month to moth or even quarter to quarter.  Some people may be waiting for the bottom of the market to purchase their next home, and as a consumer it makes perfect sense to do so but if you are relying on the Case-Shiller Index to identify when the bottom is here by the time it is reflected in the score the actual bottom will have come and gone months prior.