Consumers across the US pledge the titles to their vehicles in order to obtain quick cash through title loans. Many end up trapped in debt, and if they default they can lose their car.
The industry has thrived in Georgia, home to the nation’s largest title lender, TitleMax. A loophole in state law exempts the industry from usury laws and oversight that other lenders operate under.
Our investigation — based on an analysis of storefront locations and vehicle lien records, among others — offered for the first time a window into the scale of the industry, and the impact on their customers.
The reporting was republished across Georgia, in the state’s largest newspaper The Atlanta Journal-Constitution, as well as by public broadcaster Georgia Public Radio.
The [opening story](https://www.propublica.org/article/title-lenders-trap-georgia-residents-in-debt) was read on Apple News and ProPublica’s website by a typically sized audience, compared to similar investigations. The follow-up service journalism piece, “[How Title Lending Works,” had a much wider reach. It was read by about three times as many people as the main story. ](https://www.propublica.org/article/how-title-loans-work)
Within days of publication, seven legislators reached out to Margaret Coker asking for more information about the issue. Members of the Georgia Black Legislative Caucus from the state House of Representatives told Coker they would seek to introduce reform legislation during the 2023 legislative session, as did Sen. Chuck Hufstetler, who chairs the Senate Finance Committee.
Our biggest challenge in this project was the lack of readily available data, since title lenders in Georgia are very loosely regulated. As a result, our project ended up combining a variety of different data sources using manual and programmatic techniques.
We compiled a list of title lenders in Georgia using information from Google Maps and corporate websites. This involved a combination of manual searches as well as using the Google Maps API and using Python to scrape company websites. We also called hundreds of stores to verify that locations were in business and offered title pawns, coordinating our findings within the reporting team using Google Sheets.
Using Python, we joined our list of title lenders with census data to analyze the geographic distribution of title lenders. We found that title lenders were disproportionately located in ZIP codes with lower incomes and higher proportions of people of color. We visualized these disparities using Datawrapper.
In addition, we filed a public records request with the Georgia Department of Revenue for all vehicle title liens issued since July 2019. Using Google Sheets and a BigQuery database, we then cross-referenced the lienholders in that data with our list of title lenders to identify and analyze the liens associated with title pawns.
Context about the project:
From a data perspective, title lending in Georgia is a black box, even to policymakers. Title lenders are not licensed by the state, and, as a result, simply compiling a list of title lenders in Georgia was a significant undertaking.
We found one list compiled by an advocacy group. However, it was outdated and was missing some stores and included others that did not actually offer title pawns.
As we produced our own list, we had to be conscious of keeping our definitions and data consistent. For example, title lenders in Georgia operate under pawnshop statutes, but there are many pawnshops that don’t offer title pawns — we identified and excluded those, typically by calling and confirming that they didn’t offer such products. We wanted the product of this hard work to be available to Georgians – we put the full list of store locations on ProPublica’s Data Store.
When it came to getting information on actual title pawn transactions, we had to get creative. Title lenders don’t report any data on their business volume to the state, so we needed another data source. We discovered that there was one record that made its way up to the state level when a title pawn was issued — the lien on the vehicle title, which gives the lender the right to repossess the car if the money is not paid back.
Every lien is ultimately registered with the state’s Department of Revenue. However, as far as we knew, nobody else had used title lien data in order to size up the title-lending industry, so we were in uncharted territory.
It was challenging just to find out whether the department had the lien data in a format we could use. Acquiring and analyzing the data ultimately took months, including multiple public records requests with a total cost of more than $1,400. We also had repeated phone calls and email exchanges with the department’s IT team to ensure that we received the appropriate data and were interpreting it correctly.
What can other journalists learn from this project?
Just because data isn’t readily available on a topic, don’t be immediately discouraged. There may be a potential database you can create yourself — such as our compiled list of title lenders. Or there could be an alternative data source that doesn’t track the exact thing you’re looking for, but still allows you to reach meaningful conclusions — such as how we used vehicle liens as a proxy for counting title pawns. The process may be more tedious than finding the perfect dataset, but the product can be just as valuable or even more so, since it shines a light on what isn’t being tracked.
In addition, it’s important to remember that the data aspect of a story doesn’t exist in a vacuum, but works in tandem with other avenues of reporting. When we hit a wall on how to get numbers on the volume of title pawns statewide, we learned about the lien filing process from a source who had worked within the title-lending industry, which led us to a new data source that ultimately allowed us to evaluate the scale of the industry.