Characteristics of a Sufficient Statistic to Measure City Housing Prices

This National Bureau of Economic Research working paper, authored by Daniel Broxterman, William Larson, and Anthony Yezer, addresses the complexities of characterizing housing price levels and changes in urban areas. The study highlights challenges arising from housing stock heterogeneity, the joint consumption of housing and neighborhood amenities, and varying accessibility. Focusing on intra-city location as a key determinant of prices based on economic theory, the authors argue that existing metrics often fall short. They conclude that a sufficient statistic for describing average housing prices must aggregate data from relatively homogeneous market areas. Furthermore, these prices should be weighted by housing quantities, such as the number of dwelling units or interior space. The paper notes that common repeat-sales and hedonic indexes generally lack this specific weighting structure but suggests they could be modified to incorporate it. This research contributes to the field of regional and urban economics by proposing a more accurate methodological framework for measuring housing markets, which is central to many empirical economic questions regarding rents and asset values.