For eight years, the Tampa Police Department used the threat of eviction as a tactic to reduce crime in some of the city’s poorest neighborhoods, urging landlords to evict tenants based solely on arrests.
Officers gave landlords legal cover to evict anyone found to be involved in criminal activity, then sent landlords arrest notices and called on them to take action.
The program went under the radar until a Tampa Bay Times investigation showed that entire families were evicted based on the arrest of a single member and that 90 percent of those reported to their landlords were Black.
When the Times first contacted the Tampa Police Department regarding the program in 2017, they immediately changed some aspects of the program. The department removed wording from the notifications of arrests that stated that landlords were “required to take immediate actions through notice to cure, notice to vacate, or eviction.”. The department also sent out fewer letters in years following that first contact.
Civil rights groups — including the NAACP and ACLU — sent a letter calling for the city to “immediately end” the Crime-Free Multi Housing Program. The letter, sent two days after our first story published, cited our reporting and noted that “the Program compounds the over-policing of people of color in Tampa and causes catastrophic consequences for people of color.” On the same day, local state and city representatives called on the mayor to suspend the program for re-evaluation.
The next day Tampa mayor Jane Castor announced reforms to the program, adding police oversight and restricting the type and location of arrests that would be reported to landlords. In December 2021, the city scrapped the program and announced that it would no longer send reports to landlords detailing the arrests of their tenants.
Within a month after our story first published, the Orlando Police Department and Orange County Sheriff’s Office announced that they planned to end their own programs similar to the Crime-Free Multi Housing Program.
Our primary challenge in this project was cleaning and consolidating data from various data sources. We primarily relied on R for data cleaning and consolidation tasks.
Many of the documents provided by the Tampa Police Department existed only in PDF form. We used R’ to scrape, clean and consolidate information from the irregularly formatted PDF forms.
The datasets provided by TPD contained variations and misspellings of names, addresses, and other identifying characteristics. We used string distance tools in R and OpenRefine to clean and correct misspellings and variations.
Our next challenge was to determine whether the TPD was primarily targeting Black and Hispanic neighborhoods in Tampa. We used OpenStreetMap and various R mapping packages to locate participating buildings and pull in census data from the surrounding census block.
We found that participating buildings were in predominantly Black and Hispanic neighborhoods. The vast majority of residents evicted under the TPD program were also Black.
Finally, we used R to create a map of participating buildings compared to the surrounding demographics, which was printed alongside the story.
What was the hardest part of this project?
Tampa police used different versions of the notices they sent to landlords and also used an offender database to record arrests and evictions making it a challenge to automate the collation of more than 1,000 interactions with landlords.
Our first challenge was simply combining the various spreadsheets and filings into a once usable database of arrests and evictions. Cleaning the various formatting changes, duplicates, errors and typos took weeks of careful work.
Once we had a working database, we needed to match the names of those reported to landlords with court records and eviction cases mostly in the name of the leaseholder. This was an arduous task that took hundreds of hours to complete.
Locating and reaching out to victims proved to be a logistical challenge. Many evicted individuals remained housing insecure for years following their eviction.
What can others learn from this project?
Other newsrooms have written about similar programs in their communities but stopped short of doing a detailed analysis the Times undertook. That enabled us to report with authority rather than relying on anecdotal examples. We were able to definitely state how many tenants had been reported to their landlord for misdemeanor arrests or those that happened far from their landlord’s property. We were able to show how police targeted this initiative in mostly minority neighborhoods.