The Historical Housing Prices Project
US price and rent data from 1890 to 2006, based on millions of newspaper ads
The Price of Housing in the United States, 1890–2006. Ronan C. Lyons, Allison Shertzer, Rowena Gray and David Agorastos, June 2024.
Commentary: 101 things we now know about US housing markets (Part I), by Ronan C. Lyons.
How did the U.S. housing market perform over the twentieth century? It may come as a surprise to some but… we don’t know. Detailed information on sale prices by city only date from the 1970s while series on rents by city often date only from the 21st century (if at all).
Until now. As of June 2024, the new Historical Housing Prices Project, hosted by the Federal Reserve Bank of Philadelphia, is live. This is a project I’ve been working on for almost a decade, with my collaborators Allison Shertzer and Rowena Gray
Up to now, the long-term picture was based on work by Robert Shiller:
Based on millions of digitized newspaper ads, the Historical Housing Prices Project has been able to reconstruct a more detailed view of long-term changes in prices and rents, up to 2006 and the height of the US housing boom. (Around 2006 is when ads started shifting to the Internet.)
Abstract for the paper:
We construct the first consistent market rent and home sales price series for American cities across the 20th century using millions of newspaper real estate listings. Our findings revise several stylized facts about U.S. housing markets. Real market rents did not fall during the 20th century for most cities. Instead, real rental price levels increased by about 20% from 1890 to 2006. There was also greater growth in real housing sales prices from 1965 to 1995 than is commonly understood. Using these series we document several new facts about housing markets. The return to homeownership has varied considerably across cities and over time, but rental returns were historically much more important than capital gains in every city. We discuss the implications of our indices for the business cycle and the consumer price index. Finally, we provide evidence that housing prices increased unevenly across cities over time in response to natural building and regulatory constraints.
Capital gains versus rental returns, i.e. either renting out a home or living in it and not having to pay rent:
Variation across cities and correlation with zoning restrictions:
There is almost no relationship between real housing price growth and the restrictiveness of zoning in the 1890-1929 period. Zoning was adopted by almost every city in our sample during the 1920s. We see a slightly steeper gradient over the next two periods (1930-1945 and 1945-1980). In these periods it is possible both that the existing zoning regimes were causing higher price growth and that home price appreciation was incentivizing cities to adopt even more restrictive measures, particularly by the 1970s. The gradient in the final period (1980-2006) is even steeper, however, suggesting a closer relationship between zoning and home price appreciation towards the end of the 20th century.
The fundamental endogeneity here is difficult to unpack without better information on historical zoning regimes. However, that land use regulation has become a more important correlate of across-city differences in housing price growth over time is consistent with recent scholarship.