Inferring the Distribution of Households' Duration of Residence From Data on Current Residence Time

Abstract

Estimates of households' expected duration of residence are important to private and public decisionmakers. The common methods of estimation have been shown to be unreliable. This article presents a measurement framework for estimating total time of residence using five large sets of published government census data on the housing market. By developing a moving-behavior model, the distribution of total residence duration can be estimated from the census data on the age of current residency (i.e., time since moving into current residence). Among other results, the authors found that the average total residence duration for all U.S. households, 5.5 years, is about half the average age residence time, 10.7 years. This extended intertemporal model provides more reliable estimates for the age and expected duration of occupancy. Therefore, the model better explains and predicts housing-market behavior and also the demand for the many housing-related products and services.