What is Geographic Risk and Why Should I Care?
Certain geographic regions contain active fraud rings that find easy money through synthetic identities, stolen identities, or straw borrowers. These fraud rings are often found in large cities in the US. They fraudulently purchase vehicles on credit and either sub-lease or sell those vehicles overseas without ever making a payment.
But, as every lender knows, it is illegal to differentiate risk based on neighborhoods or other narrow geographic definitions. Redlining, as this practice is called, prevents discrimination against minority groups that often live in high-risk communities. It protects the financial interests of hard-working individuals who deserve fair interest rates. After all, they and shouldn’t be excluded from opportunities based on where they live.
How Can Lenders Prevent Geographic Fraud Risk Losses?
Beyond violating the discriminatory practice of redlining, what can lenders do to prevent geographically-based losses?
Many lenders already leverage identity-related fraud prevention tools like Point Predictive’s SyntheticID and Auto Fraud Manager to reduce their risk. But they struggle to differentiate the synthetic identities created by fraudsters versus those created by individuals fully intent on paying off their loans. And straw borrowers can be even more difficult to detect.
Fraudbot helps solve these issues by constantly scraping Point Predictive’s data consortium for patterns indicating increased fraud ring activity, which is where the biggest losses occur. Then this powerful machine-learning tool determines how to identify fraud ring activity in real-time. As a result, lenders fund fewer fraudulent loans.
What Kinds of Patterns Does a Fraudbot Find?
Fraud Ring Cycles: Fraud rings typically cycle through periods of high activity and slumps of low or inactivity. FraudBots pick up on these cycles in different areas to alert lenders when the risk of fraud is at its highest.
Top Vehicles Most Commonly Involved in Fraudulent Auto Loan Applications: Typically, fraudsters look for the vehicles that will net them the most profit. They tend to target specific models over and over again. Fraudbots watch these trends and alerts lenders when a suspicious borrower tries to purchase one of those vehicles.
Misrepresentation: Often, different fraudsters working with the same organization use the same identifying characteristics. The same home address, email address, phone number, and employer show up across different identities. Fraudbots pick up on these and notify lenders when known fraudulent characteristics appear in their applications.
For example, in Houston, Fraudbots found a home address used by 605 different “applicants” in the past few years. Many addresses are associated with many identities. However, this find enabled Point Predictive to flag hundreds of fraudulent identities and individuals. Naturally this single linkage pattern saved a substantial amount of fraud across many lenders.
How FraudBots Help Lenders
By looking at data across many lenders, Fraudbot picks up on fraud ring trends lenders would never be able to see. Since Fraudbot was designed to accurately identify trustworthy data, honest customers are able to see their loans funded faster, and with fewer documentation stipulations. Fraudbot accurately warns lenders prior to funding about potential fraud losses when it identifies new bad actors. Fraudbot alerts lenders quickly if they have already funded a loan involving a recognized pattern.
While it’s no surprise that fraud would be more prolific in large cities, Fraudbot’s heat map has uncovered fraud ring activity in these geographic regions:
Learn more about how Point Predictive’s Artificial and Natural Intelligence [AI+NI]TM approach can help reduce your geographic fraud risk without redlining.