Historical Local Housing Supply and the Origins of the Affordability Crisis
This project investigates the long-run origins of the modern U.S. housing affordability crisis by constructing the first comprehensive, jurisdiction-level panel of housing supply elasticities spanning most of the 20th century. Recent evidence shows that home prices have increasingly outpaced incomes, while regulatory constraints such as exclusionary zoning have proliferated since the 1970s. Yet, existing measures of supply elasticity are geographically coarse and historically short, leaving unanswered how local rules and physical constraints shaped today’s affordability challenges. We will address this gap by digitizing historical building permit records and developing new elasticity measures at the county and town levels.
Beyond measurement, the project speaks to major questions in urban and regional economics. A better understanding of historical supply constraints can clarify why affordability deteriorated, how suburbanization accelerated, and how land-use rules reshaped U.S. cities. Our work will open new research avenues on zoning, demographic change, transportation patterns, and the macroeconomic implications of persistent housing shortages.
Requisite Skills and Qualifications:
The data acquisition portion of this project involves compiling records from PDF scans of historical government reports and periodicals such as newspapers and investor manuals. This may require a combination of using optical character recognition (OCR) techniques and software such as ABBYY, as well as recent Python packages designed to convert historical texts into a machine readable format. Much of the work will involve harmonizing tables from the historical records over different time periods and across changing definitions of geographic boundaries.
Prior experience with web/text scraping would be useful. Candidates who have experience applying ArcGIS, Google Maps API, and other software packages related to geo-locating data points are especially encouraged to apply.
Proficiency in statistical/econometric software packages such as Python/R is required for aspects of the project which involve cleaning the source data collected. Otherwise, a working knowledge of Excel and attention to detail are sufficient.
I welcome applicants with interests in other fields besides economics – including but not limited to – urban studies, computer science, mathematics/statistics, physics, and political science.