Last week, Zillow.com told me my home's market value has plunged 7% in the 5.5 months since I bought it. Before I did anything like question the veracity of its conclusions, I worked through the five stages of grief -- denial ("No! This can't be!"), anger ("Stupid site! What does it know, anyway?"), bargaining ("If I only knew how it worked, I could understand this number."), despair ("Oh, God, it's true -- we bought a house at exactly the wrong time.") and acceptance ("Ah, I got Zillowed. Whatever.")
Having gone through that, here's why Zillow.com struck me as maybe not the best tool for assessing fair value: Like I said, it claims my house's value has plummeted 7% since August; it also said that housing values in my county have dropped 7%. This struck me as not entirely correct: According to the California Association of Realtors, median housing prices in Alameda county have logged double-digit year-over-year increases since we bought: Aug -- 19.2%; Sep -- 21.3%; Oct --16.6%; Nov --19.4%; Dec -- 14.9%. According to the National Association of Realtors, home prices in the San Francisco/Oakland/Fremont region rose 9.4% in Q4.
The data would seem to point to rising market values for properties in my area.
Another reason the site seems hinky to me: it's not in keeping with the sales trends I've personally observed. I've been watching home listings and sales on the island with some interest, noting the price ranges at which comparable houses are a) listed and b) sold. As of last Sunday, we were among the lowest three sales prices on the island in the last six months; prices have been trending up ever since we bought.
You can see where I was wondering how Zillow's data was so incompatible with my firsthand observations and research. I'm not the only one: Leslie Walker reports in "Site's Alluring Facade Is Built on Shaky Foundation" (WaPo, Feb 16, 06) that Zillow's valuation formula has been overriding actual sales data too frequently. This goes a long way toward explaining why our own Zestimate seemed hinky when examined in context with other data.
Walker also writes this:
Nationwide, 62 percent of all Zestimates fall within 10 percent of the selling price, according to Zillow. That means 38 percent are more than 10 percent off the mark, which strikes me as significant.
So you have a nearly two-in-five chance of your Zestimate being way off the market, and a three-in-five chance of your Zestimate being 10% over or under the selling price. Ten percent may not seem like much in the grand scheme of things, but for anyone who plans on refinancing once they hit a certain equity benchmark, 10% is the difference between PMI and a plain ol' mortgage. That 10% margin of error can be mighty pricey -- especially in a state like California where more than half of new homebuyers in the last year have the kind of nontraditional financing that's linked (directly or indirectly) to the amount of equity they'll get from their home.
It'll be interesting to see how Zillow.com affects the house-buying process -- whether it'll gain credibility, undermine the credibility of realtors and appraisers, or be dismissed as a valuable tool.
Recent Comments