Housing Impacts of Data Centers

This research looked at how data center construction and operations affect nearby communities, including housing prices, employment, local tax revenue, and various environmental indicators. I worked at Berkeley's Energy Institute at Haas, in partnership with The Opportunity Lab at Berkeley, and I developed a multi-source dataset combining data center facility location records, census tract demographic data, and Zillow housing indices to study economic impacts over time.

Data Collection: To build the dataset, I scraped facility records from public permitting databases, state utility filings, and industry sources. For example, I found undocumented API calls from Business Insider's interactive map using browser DevTools, then parsed and cleaned the raw JSON payloads into a geocoded CSV that could be combined with the other data I found. Housing price data was sourced from Zillow's ZHVI dataset and county-level Census variables were pulled via the Census API, covering demographics, income, employment, and other housing-related data.

Exploratory Analysis: The initial EDA (Exploratory Data Analysis) examined datacenter clustering by geography and construction timeline, validated that treatment and control counties followed similar pre-2021 price trends, and mapped facility density against housing price changes.

Future of the project: The end goal of the research is to create a difference-in-differences analysis comparing housing and economic outcomes in counties with data centers against matched control counties. The dataset and initial analysis I have conducted will support the future of this research.