An Analysis of Commercial Evictions in the US from 2000 to 2016

Abstract

This thesis examines the extent to which establishments are evicted in a sample of 29 US states, identifying top-evicting areas and characteristics that predict eviction frequency. In order to incorporate both spatial and temporal dimensions into the analysis, I use seventeen years of individual-level commer- cial eviction data and aggregated data at the state, county, and block group levels on residential evictions and location demographics, obtained from the Eviction Lab at Princeton University. Working with a combination of multiple years of data at di↵erent levels of geographical granularity provides method- ological advantages. Primarily, it enables a qualitative component of describing commercial eviction situations both within and across locations, in a degree of detail not always extractable from quantitative methods alone. I therefore use both descriptive findings from the data alongside results of OLS models to present a more complete narrative of commercial eviction. I find that the number of cases in a county is positively related with median gross residential rent, and that the situations of top-evicting cities of Chicago and Las Vegas indicate the substantial extent to which the forces driving commercial eviction can diverge.