Sellers Who Bought Post-Bubble More Likely to Over-Price Home

Imagine two identical houses built in the same year in the same neighborhood.  House A was last purchased in 2006 and House B in 2008.  House A is listed at its estimated fair market value of $300,000.  Although it would be logical to assume that House B would list with a similar asking price, new research shows that it would, in fact, list at $350,000 on average, a $50k premium!  Why the 16 percent price difference?

An analysis of seller behavior reveals that homeowners who bought after the peak of the national market in June 2006 dramatically over-price their homes relative to its estimated market value.  In a separate survey fielded by Zillow, 17 percent of sellers who purchased post-bubble claim that their primary factor in pricing their house is their original purchase price.  This compares with 9 percent who bought during the run-up to the bubble and 4 percent who bought before that.

In the chart below, the blue line is showing the difference between the current list price and the estimated market value of the home with the year the house was last sold running along the X axis.  The green line represents the difference between the current list price and the prior purchase price.  Notice in the green line that current sellers that purchased their home since 2009 have been pricing their house at 10% higher than what they purchased it for just 1-2 years ago.  This is in spite of the fact that over the last two years the national real estate market has depreciated by 10 percent.  This difference is represented in the blue line which shows that sellers who bought during this period are pricing around 20% above market rate. Not only are these sellers ignoring the losses they have taken since purchase, but they’re trying to claw back all of their closing costs too it seems!

Obviously the idea that your largest asset has been devalued significantly is difficult to accept, however, people who bought in the run-up to the bubble are seemingly more willing to confront this reality than those who purchased after the peak.  In fact, relative to sellers who purchased their home before 2002, those who bought while the bubble was expanding rapidly are comparatively underpriced.  When first placed on the market, the typical house is priced at roughly 10 percent above its estimated market value, but sellers from 2006 touch as low as 6.4 percent.  Looking at sellers who bought on either side of the market peak nationally reveals stark differences between these two groups.  Sellers who bought in January 2006 overprice their home by only 8 percent, while those who bought in January 2009 overprice by 22 percent.

Sellers who bought post-bubble seem to think that since their home purchase occurred after the peak of the market, and thus home values were already significantly discounted relative to the peak, the seller escaped the worst of the bubble.  The problem is that “The Bubble” didn’t pop so much as steadily deflate for the better part of 5 years now, and current home values now represent what they were worth in 2003.  Said differently, assuming your market followed the national trend, unless you bought your house before 2003, you should be selling it at a loss now.  The closer to 2006-2007 you bought, the bigger that loss should be.

We know there are a million numbers to keep in your head when looking at a potential property, and that by no means does every property purchased in 2008-2010 is dramatically overpriced.  However, I humbly suggest that when looking at properties, you keep one more very important, and very simple, statistic in mind: Previous Year of Purchase.


Zillow’s analysis was done by taking one million currently for sale homes with prior sale data since 1999 and looking at the difference between the current list price and the previous sale price.  We then compared the change in the Zillow Home Value Index of that property’s zip code from when it was previously sold to now.  These data were grouped by month and the median value, as well as the median difference between the two metrics, was then calculated.  The resulting graph and data as well as the survey information yielded the above conclusions