Historical Housing Price Data

Project Outcomes Statement

The primary aim of the project was to combine state-of-the-art methods in housing price index construction with newly digitized newspaper archives to create consistent housing price data for urban areas in the United States. The project successfully created quarterly or annual samples for 30 major cities for the 1890-2006 period, with about 2.8 million listings in total. Each digitized listing contains size, price, and location information in addition to other property attributes. This data was used to create new price series for owned and rented housing for these cities.

These series were used to ask how the return to owning a house has evolved over time and across cities in the United States. Market rents were used to revisit the question of how much shelter prices have increased over the twentieth century. These questions are central to understanding both housing’s performance as an asset and also how housing has shaped increases in living standards in the twentieth century. The project also investigated how much housing prices moved with the business cycle and asked why housing prices have increased by more in some cities compared with others. In the future, the data can also be used to study long-run trends in housing affordability across cities investigate housing bubbles in American history.