“De-anonymizing” landlords with computational tools 

Are institutional and large landlords overtaking the rental market? Acquisitions by institutional and large landlords now create a great deal of anxiety from lawmakers and the public. However, this is a difficult question to assess as property records are difficult to work with, and property owners have many tools to hide their ownership. As a result, there are no existing complete studies of ownership, and it is difficult to assess recent transformations in rental landlording or make statements about trends.

I have developed a new computational approach for solving this problem, which uses massive networks to algorithmically search for owners through property records and business filings. In Austin, Texas, a hotbed of recent investor activity, I have constructed property ownership networks from Travis County Central Appraisal District property data and business records filed with the Texas Secretary of State. These “entity networks”, which link more than 36 million persons, companies, addresses, and properties in Travis County, TX from 2010 to 2021, allow linkage of property ownership to owners across shell companies, LLC ownership structures, and myriad address and name harmonization challenges.

Initial findings from this project in Austin are available here. I aim to provide public versions of this dataset by Summer 2024.

Understanding urban housing movements and community organizing through big data

In addition to studying landlords, I am also interested in how urban housing movements respond to recent transformations in property ownership and affect renter housing insecurity. Pioneering neighborhood effects studies provide some leverage on these questions in Chicago (the PHDCN) and Los Angeles (L.A FANS), when they examine how formal organizations influence community solidarities and exposure to economic insecurity. However, the majority of research in this area is based on ethnographic and historical case studies, and few comparative or systematic studies of urban housing movements and community organizations exist.

Consequently, a key part of my research agenda has focused on bringing interorganizational networks “back-in” to urban sociology to enable scalable, comparative research on civil society coalitions and urban housing movements. The goal of this data project is to enable researchers to test fundamental questions about urban governance, activism, housing insecurity, and municipal policy change. With help from Jesse Lecy and IRSx, I have used administrative data on nonprofit organizations to construct longitudinal networks among civic leaders in the 212 largest U.S. cities. This data set, currently under finalization, provides data on relationships between 12 million civic leaders in all U.S. cities between 1998 and 2016. You can read initial findings based on this data in my article in Social Networks. Replication files and the raw data used in this analysis are currently hosted on my GitHub. Variants of this database have also been used to directly test the effect of community organizing on eviction—one of the most catastrophic experiences a family can have. In a recent article published in Social Problems, I show that the growth of pro-tenant community organizations in large U.S. cities had a meaningful causal effect in reducing eviction filings after the Great Recession.