Eviction Statistics are Tricky, so Be Careful of First Impressions

Eviction Labs. Chart of cities with highest eviction rates.

Detroit    I was scratching my head.  Everything the ACORN Home Savers Campaign knew from our experiences in building committees in cities around the country and talking to renters and land purchase contract signers indicated that evictions and the threat of evictions were soaring.  In fact, in no small measure that was what was driving families into the hands of often predatory contracts.  Yet, this week looking at the national database from the Eviction Lab directed by Matthew Desmond, now a Princeton professor, but earlier well-known for his powerful book, Evicted, about life for lower income families desperate for affordable housing in Milwaukee, it almost seemed to indicate the opposite of what we were seeing in the field.  What’s up with this?

First, let’s look at the huge database from the Eviction Lab.  Many states are actually included on their site, like Texas and Louisiana for example, that were excluded in the mapping report in The New York Times.  What’s surprising is the charting.  Over and over again, state by state, and nationally, the trend line on evictions decreases after a peak in the several years after the real estate bubble popped in 2008 and foreclosures were mammoth.  Furthermore, these numbers are mirrored in the Eviction Lab’s reporting on state levels as well.  If you dig deeper and look at specific cities, they are still trending higher, but the data is an aggregate, so the horror of evictions is actually diluted on average as the geography expands.

Perhaps that is statistically true, but it isn’t really helpful or a reflection of reality.  The Times map was in some ways a correction, because it showed county-level numbers that were rising or falling.  The actual Eviction Lab numbers though would seem to arm the industry and landlords with an argument that, “hey, it’s no big deal.”

This is part of the false truth of big data.  Only the numbers that are officially reported can be crunched.  Most evictions and apartment abandonment are never reported.  The Census Bureau only agreed in 2017 to begin assembling counts that could create a national database.  A similar problem exists in the shadowy world of land contracts where there is also little accurate data and less accurate reporting, and where the Census Bureau inexplicably stopped counting during the Obama years.  There is every indication that the reported numbers are only the tip of the iceberg.  Landlords and Desmond know this regardless of the Eviction Lab database.

City Lab for example in a report six months ago found what we are seeing.  As they indicate,

“A new report from Apartment List aims to more accurately estimate the scope of the population at risk of eviction, building on data from its 8 million users, plus answers to 41,000 surveys on rental security. The scope, they found, is wide, and growing: One in five renters recently struggled or were unable to pay their rent, and 3.7 million renters nationwide have experienced an eviction in their lifetime as a renter.”

City Lab quotes from Desmond’s own 2016 work saying,

According to data collected in Milwaukee by Matthew Desmond, author of the Pulitzer-winning 2016 book Evicted, more than one in eight renters in that city experienced a forced move between 2009 and 2011. Only 24 percent of them were due to formal evictions. Another 48 percent were informal—instances where the landlords offered money for tenants to leave, or they removed doors from their hinges, or the tenants themselves got up and left.

If the same ratios held that would indicate only about one-quarter of evictions were formal, and it is still likely higher, especially in the South where even the Eviction Lab found rising levels and reporting is laxer and incomes even lower, then the national eviction rates would be closer to 12-13% than the 3+% that the Eviction Lab was reporting on formal evictions nationally.  Let the data be a wakeup call rather than a guide.  Evictions are soaring especially in lower income and racially diverse and minority areas.  This is a national crisis, so the numbers can’t be massaged to make them look any better, especially while so many are fighting for real solutions for tenants.

Facebooktwittergoogle_plusredditpinterestlinkedinmail