Loss Aversion in the Housing Market: A New Paradigm

We propose a search model in a housing market which incorporates Tversky and Kahneman’s prospect value function as a special case. Our model provides a clear correspondence between each component of the prospect theory and its unique empirical implication. Overall,under the prospect value function, we predict a non-linear, non-monotonic pricing curve with changing curvature. One striking prediction from our model is that marginally diminishing sensitivity may lead to a reverse disposition effect. That is, when a prospect-utility seller is subject to a range of moderately sized losses, her asking price can be decreasing when the potential loss is increasing. These findings therefore suggest that the classic test of loss aversion in housing markets, as led by Genesove and Mayer [2001], may be invalid. Furthermore, we predict a larger price dispersion in a cold market than in a hot market and also reaffirm the positive price-volume relation. Using single family transaction data from Hampton Roads in Virginia from 1993 to 2013, we find a close match between the empirical findings and our model’s predictions. Finally, the steeper slope of the empirical pricing curve suggests that the marginal diminishing sensitivity, a key component of the prospect theory, could be changing based on the loss/gain position of the agent.