The Cracks in the Eviction Lab Wall are Undercounting the Crisis

ACORN Citizen Wealth Financial Justice Foreclosure
Facebooktwitterredditlinkedin
Anti-Eviction Mapping Project

New Orleans   When Princeton’s Eviction Lab came out with their national report on many jurisdictions court records of evictions, we had positive things to say about the project, but we were skeptical.  The report seemed to undercount the experience in the community of ACORN Home Savers Campaign.  A critical flaw seemed to be the very fact that only actual adjudicated procedures were part of the dataset, ignoring the vast number of informal evictions and pressured departures that are more common than court filings.

I was scratching my head, as I noted then, because Matthew Desmond, the prize-winning author of Evicted and founder of the Eviction Lab knew better.  In his book he had used a small informal survey of forced and informal evictions that was many times higher than what he was now reporting.  The actual stories from Milwaukee that were the backbone of the book’s attraction also highlighted the ways that high rent and bad conditions kept lower income tenants constantly moving between tenancy and homelessness in order to keep one step ahead of the landlord.

An article in Shelterforce by eight veterans of tenant and housing activism along with extensive experience in data collection called “Eviction Lab Misses the Mark” details that there was a bonfire beneath the smoke I was smelling.  Their point-by-point critique is so devastating that it calls into question whether the data from Eviction Lab is a tool for landlords and policy apologists, rather than a step forward in looking at the impact of soaring evictions on impoverished communities.

Among the points the authors make are these:

  • The Lab resisted collaboration with established data centers from housing activist groups by not sharing policy goals, protocols to protect tenants and confidentiality of data, and offering fair attribution for assistance, much less being willing to pay for the time and trouble of acquisition of the data sets to other nonprofits.
  • The Lab instead in known cases contracted with private business groups involved in “tenant screening” for landlords paying one such group in California over $100,000 to assemble data for the Lab.
  • Where data had been rigorously collected in the community in places like San Francisco, Eviction Lab’s figures undercounted the reality by 50%!
  • Where the Eviction Lab performs a service by trying to establish a national baseline, they fall short in falling to integrate local regulations that impact and potentially distort their numbers, retarding policy changes for the poor, rather than advancing them.

The authors tell one horrific story where tenant groups and others paid $20,000 for Desmond to come to Portland, Oregon to speak on evictions.  The groups were trying to advance a campaign they were waging to replace no-cause evictions in favor a just-cause eviction policy.  They had already won some relief in extending notice up to 90 days even under existing policy.  Because Oregon allows no-cause evictions without court filings, the Eviction Lab figures were drastically undercounting the reality on the ground.  Desmond refused to listen to explanations of flaws in the Lab’s numbers when he was there, and in defending the erroneous numbers from the lab, actually complicated the authorities saying that the fabricated numbers proved they “were doing something right” and naysaying whether no-cause evictions were even a problem.

Hey, anyone can flub up and miss the nuances of local situations when they are parachuted into town, but not even asking or listening, and then weighing in as an expert on something as serious as local policy prescriptions and housing conditions for lower-and-moderate income tenants facing fierce forces of gentrification, that’s a level of hubris that discolors not just the inadequate data and summary of the Eviction Lab, but the entire project.

Desmond and the Eviction Lab need to either get this right or get as far away from this field as possible.

Facebooktwitterredditlinkedin