Berkeley Program on Housing and Urban Policy

Estimates of the prices of housing and the value of the stock are derived from observations on housing transactions. These transactions may well be a non-random sample of the underlying population of dwellings. For example, it is widely thought that smaller "starter homes" sell more frequently than more expensive properties and that the frequency of transactions on high-valued properties varies over the business cycle. This paper considers the importance of these selectivity issues in making imputations about housing price trends. We estimate a model of housing price determination and of the nonrandom selection of observed transactions. We analyze the factors affecting the probabilities that transactions on different houses will be observed, and we estimate the effect of these factors upon housing prices. The analysis considers a variety of plausible selection models. For each of the alternatives, the estimated effect of selectivity upon housing price calculations is quite substantial.

The analysis is based on a unique body of data containing observations of all house sales in Sweden during the period 1981- 1993.