One of the differences between the social sciences and the physical sciences is that the social sciences tend to be comforting whereas the physical sciences are often disturbing. The physical sciences often conflict with our world views -- things aren't really as they appear to be, what we thought was good for us turns out, really, to be bad for us, and vice versa. Whereas the social sciences frequently confirm what we believed -- "divorced people don't get along with each other," "most interviewed would rather be rich than poor", etc. The application of social science investigations to real estate, however, yields mixed results. Some are intuitive, some are surprising, and some are just plain silly.
Three different researchers have looked at the relationship between the remarks made in MLS listings and the sales prices and time-on-market of the listings that are described. Ronald Rutherford et al. examined 58,386 residential transactions sold and closed through the Metroplex MLS in Tarrant County, Texas, between January 1994 and December 1997. Paul Anglin looked at more than 20,000 listings provided by the Windsor, Canada, real estate board for the years 1997 to 2000. The largest study was conducted by Steven Levitt and his associate, Chad Syverson, in which they analyzed more than 98,000 sales conducted through the Multiple Listing Service of Northern Illinois during the period 1992-2002.
Many of the results are what we might have expected. Rutherford notes, for example, that many of the remarks that would be deemed as positive (e.g. "on golf course", "near lake", "updated") are associated with listings that sold for higher than average prices. Moreover, comments that describe negative features ("needs repairs", "foreclosure") are correlated with decidedly lower selling prices.
Other cases are not so clear. Anglin found that in MLS listings where the description said "seller motivated", the property actually stayed on the market 30 percent longer than did typical properties. This is similar, though a bit more dramatic, than Rutherford's finding that "seller motivated" properties experienced a 15 percent longer marketing period than did the average. Neither had an explanation for this. Nor do I.
Other results range from puzzling to inconsistent. Anglin and Rutherford both found that properties that were described as "in good condition" or "beautiful" sold more quickly and at a higher price than did the average comparables. Yet Levitt's data revealed that homes described as "fantastic" or "charming" actually sold for less than did average comparables.
The attempts to explain such anomalies are revealing. Rutherford and Anglin take the somewhat common-sense view that remarks don't cause the differences, they simply convey information about the property. It is the reality behind the remark that explains the result. Levitt, on the other hand, is a conspiracy theorist extraordinaire, who sees the words occurring in the MLS remarks section as much more fraught with meaning than being simple descriptors of the property. For example, "fantastic", he says, "is a dangerously ambiguous adjective, as is 'charming.' Both these words seem to be real-estate agent code for a house that doesn't have many specific attributes worth describing."
Levitt loves the idea of there being a real estate agent's "code." Thus he explains the anomalous finding, in his data, that "well maintained" is correlated with lower sales prices. "This doesn't necessarily mean that labeling a house 'well maintained' causes it to sell for less than an equivalent house. It does, however, indicate that when a real-estate agent labels a house 'well maintained', she is subtly encouraging a buyer to bid low." (From Freakonomics, by Steven Levitt and Stephen Dubner) This comment, of course, assumes that MLS remarks are directed to buyers; and, presumably, it also assumes that buyers have been able to learn the "code".
One thing you can say about the social sciences -- at least when they are applied to real estate -- that you are not liable to hear about the physical sciences: their 'findings' can be entertaining.
