In my last post I discuss various statistical methods used to track housing prices. I suggest that the Case-Shiller Index, which uses the repeat-sales method, is the current gold standard.
Case-Shiller's utility is limited by the relatively few metro areas it covers--twenty to be precise. I happen to live and work in the Harrisburg, PA metro area. The nearest city included in the Case-Shiller Index is Washington, DC, one hundred miles south and a far different housing market than Harrisburg (trust me).
So...if you live in a smaller metro area you probably depend on your local multi-list service to periodically report sales statistics, usually the mean (average) and median home price. Personally I like to present this data to home sellers when I talk about the market they are selling into. As my local multi-list reports both the mean and median, I wondered which stat better represents central tendency for my marketplace.
To this end I collected a year's worth of home sales data for the West Shore of the Harrisburg metro area (the collection of munipalities west of the Susquehanna River). For 2009 there were 1,962 residential units sold. The histogram below shows the distibution of home sales. It has a Bell Curvish shape skewed to the high end. Thirteen sales over $700,000 largely account for the long "tail" on the distribution. I suspect this is a characteristic of many real estate markets that include very high-end communities.
Superimpose the mean and median on this distribution (see below) and you "see" a $20,000 gap between the two measures of central tendency. Visually the median appears closer than the mean to what most people would consider the typical house price. Also, in this case, the median corresponds with the mode, which is the most frequent value in a distribution.
The conclusion of my simple test is that the median seems the more appropriate measure of central tendency or typical-ness in the housing market. As it is less sensitive than the mean to outlying values, i.e., the relatively few high-priced home sales, it should also be less volatile quarter-to-quarter or year-to-year, and therefore a better statistic with which to trend housing prices.
John,
ReplyDeleteI don't disagree with the statistics but to say that median price is a measure of central tendency of a market seems to tell only part of the picture to me. Your monitoring price, but what does that price buy? Is it a 3000 sf. 5BR/4BA house or a 1200 sf. 3/1? In my opinion, to truly value a market I think you need to break it down my segments. As we've seen, the first time home buyers proportionally flooded the market during the first quarter of 2010 due to the tax credits. That would artificially skew the statistics regarding the number of units sold as well as the price point of the units being sold.
First-time buyers cannot "artifically skew" the market, they are the market. Still, I get your point...if you're listing a $450K property and you think a recent government policy preferentially stimulates home sales below $200K, you may want to exclude under $200K homes from the sales stats you discuss with your high-end homeowner. Common sense applies...best thing is probably to show both sales stats side-by-side.
ReplyDelete