Search, matching, chains and house prices

Abstract The housing market is an asset market with some unusual features, as agents have idiosyncratic preferences for housing so that search is required, and all parties must make matches before a transaction can be¯nalised. We present a simple model that re°ects these features and¯nd that the search parametersprobability of matching, for example -have the expected e®ect on asset prices, but of second order magnitude. Large falls in the probability of matching (and hence long increases in mismatched duration) are likely to have small e®ects on house prices. This helps to explain the observed persistence in house prices in the face of large changes in turnover. We also show that the number of new entrants to the housing market may have a large in°uence on the probability of selling a property