Fraud in auto lending has never stood still and neither can we. 

While synthetic identities, doctored paystubs, and misrepresented employment have long challenged the industry, a new method is quietly gaining traction: circular income transactions on digital bank statements. This isn’t just fraud, it’s evolved, digitally enabled, and harder to catch without the right tools.

Here’s how it works: A consumer inflates their income by initiating multiple cash transfers via peer-to-peer platforms like Venmo or Cash App. These funds hit their checking account just in time for the statement period, giving the illusion of recurring deposits. Before lenders or systems can detect the pattern, the funds are moved out again, often to a different personal account, a friend, or simply back through the same app. 

It’s synthetic income laundering in a closed loop — circular transactions that create a false baseline of earnings. 

Real transactions, real platforms 

What makes this tactic particularly dangerous is its subtlety. Unlike altered PDFs or fake employers, these are real transactions on real statements, processed through real platforms. Automated income verification tools are not equipped to detect the timing, pattern and reversals of these flows, and may be tricked into flagging the borrower as “qualified,” thereby putting the lender at unnecessary risk. 

The trend is emerging across credit tiers and geographies, driven by pressure on affordability. As vehicle prices, interest rates and insurance costs rise, borrowers are looking for ways to make their applications work on paper. And digital financial tools, while convenient, have become the fraudster’s new playground. 

Rethinking ‘consortiums’ to tackle fraud 

The industry often turns to fraud consortiums to surface these kinds of patterns. But the term triggers red flags from privacy, compliance and legal teams. With data-sharing boundaries governed by the Gramm-Leach-Bliley Act (GLBA), many lenders hesitate to participate, fearing the perception of data misuse or competitive exposure. 

The answer isn’t to step back, but to modernize the framework. Rather than framing these efforts as traditional consortiums, we must pivot to “fraud intelligence exchanges,” for example. These are privacy-compliant, opt-in environments designed to identify behavioral anomalies without exposing consumer-level data. Think: tokenized intelligence rather than raw personally identifiable information, or consider outcome-based flags rather than shared customer records.   

In the case of circular transactions, the key signal isn’t the amount, it’s the velocity and reversibility of funds. That’s the kind of insight that consortium-alternative models can deliver, especially when powered by AI models trained to detect digital behaviors, not just documents. 

Multiple detection layers 

To combat these new forms of fraud, it is best to rely on a system that identifies fraud through multiple layers of detection that go beyond what human reviewers or surface-level automation can catch.   

Anomalous collision flags unique identifiers, such as employee code, transaction ID, or anomalies that are highly unlikely in addition to phone numbers or email addresses, that appear across multiple documents in our database, suggesting recycled or shared data used across different applications.

Fraudulent template detection compares incoming documents to known fraudulent layouts and formats, helping us identify reused templates that have already been tied to fraud events.

Metadata analysis surfaces cases in which documents have been edited multiple times or show signs of tampering, even when the visual content appears clean.

Typography and formatting checks catch inconsistencies in font, spacing, layout, and alignment that deviate from authentic documents issued by verified sources.

A multilayer approach works because fraud is no longer just about fake documents, but includes behavioral manipulation embedded within digital records. Circular transactions are a perfect example. The deposits may be real, but the intent behind them is not. Without tools that analyze behavioral context and structural anomalies, these schemes can slip through. 

Looking ahead 

This isn’t just a compliance issue. It’s a profitability issue. Fraudulent approvals lead to early defaults, charge-offs and repurchase pressure. Worse, they undermine trust in automation, slowing digital transformation and hurting the borrower experience for legitimate applicants. 

Lenders must invest in tools that go beyond surface-level statement reading. That includes partnering with vendors capable of parsing raw bank data, analyzing transaction flows over time, and flagging anomalies in real-time, without waiting for a loss event to trigger a review. 

As we face down the next wave of fraud, we must recognize that synthetic income isn’t just fake paystubs anymore. It’s hidden in the flow of real cash, driven by intent, and dressed in digital legitimacy. 

And if we’re going to stop it, we’ll need to think as creatively as the fraudsters themselves. 

Jessica Gonzalez is the vice president of customer success at Informed.IQ and has more than 15 years’ experience in the financial services industry, including tenures at Santander Consumer USA and Visa.  

Content sponsored by Informed.IQ 

Auto Finance Summit, the premier industry event for auto lending and leasing, will host a fighting fraud panel with Dawn Carpenter, chief risk officer of Mercedes-Benz Financial Services USA, Megan Prouty, AVP of consumer lending operations for PenFed Credit Union, and Anthony Capizzano, SVP and head of consumer lending for Axos Bank. Auto Finance Summit returns Oct. 15-17 at the Bellagio Las Vegas. Learn more about the 2025 event and take advantage of the Aug. 29 early-bird registration here.