A Flexible Method of House Price Index Construction Using Repeat-Sales Aggregates

The major issue which we address in this paper is the one-size-fits-all nature of the typical city-level house price index. In this vein, we make two contributions. First, we develop a new algorithm to ensure feasible estimation of geographically granular repeat-sales house price indices in cases of low transactions counts. This facilitates the estimation of a balanced panel of 63,122 Census tract-level repeat-sales house price indices (2010 definitions) at an annual frequency between 1989 and 2021, which we release alongside this paper. Second, we use these indices to estimate city-level house price indices that are robust to heterogeneous submarket appreciation and non-random sampling, two issues that confound classic approaches. Numerical simulations show this algorithm uncovers population indices even when house prices, quantities, and transaction sampling vary across locations and over time. This approach can be used in a flexible manner to calculate canonical price indices such as Lowe and Laspeyres, and more tailored summary indices on a variety of topics, including collateral valuation, climate risk assessment, or tracking changes to minority housing wealth over time