Morris--along with Erwan Quintin--calculates median house prices by MSA using the American Community Survey from 2006-2010. Because the ACS samples all houses, the change in price from year to year is largely not biased by the change in composition of the housing stock (the only change comes via new construction and home improvements--and the US had little of either from 2008-2010). As such, the calculation, which is based on what people think their house is worth, is in some ways superior to house price indexes, which inevitably suffer from composition bias, even when their designers make admirable efforts to mitigate such bias.
In his talk, Morris showed that people thought the value of their houses went down substantially less than Case-Shiller implies. Where Case-Shiller or people are right is not particularly important to mortgage performance, because people will not default if they think their house is worth more than their house. Those who are forced to move for economic reasons might find themselves unpleasantly surprised, and may wind up selling (now) through a short-sale. But it is possible that the reason many underwater borrowers are not walking away is that they think they are not under water.
In his talk, Morris showed that people thought the value of their houses went down substantially less than Case-Shiller implies. Where Case-Shiller or people are right is not particularly important to mortgage performance, because people will not default if they think their house is worth more than their house. Those who are forced to move for economic reasons might find themselves unpleasantly surprised, and may wind up selling (now) through a short-sale. But it is possible that the reason many underwater borrowers are not walking away is that they think they are not under water.
