To say the least, 2020 was by far a surprising year that made us go through several eye-opening situations. From the toilet paper shortages to statewide stay-at-home orders, we’ve witnessed firsthand what this pandemic can do to a variety of industries adjusting to the new norm. There have been several changes of behavior, and it’s become harder to define what behavior is considered “normal” and what’s considered not.

Here we’ll take a look back at what specific fraud trends banks had to look out for in 2020:

1. Frankenstein Identities

Have you heard of synthetic identity fraud? It’s the fastest-growing financial fraud scheme, and it’s also one of the hardest to detect. Rather than stealing another person’s identity, a fraudster invents an entirely new and fake persona. They buy a random social security number on the dark web and link it to a fake name, date of birth, and sometimes social media accounts. They typically look for a social security number issued to someone younger than 18 years old so it seems like that person hasn’t established a credit history yet. They then start applying for credit and building up the fake account’s credit score. While this is no easy feat, because it can take years to accomplish, the goal of these fraudsters is to obtain an 800 or higher credit score so being approved for multiple high limit credit cards and unsecured loans from banks is easy to do.

You’d think banks would be able to flag these situations easily, but that’s not the case. Due to human error or a legal name change, it’s not uncommon for social security numbers to be linked to multiple names. So how can banks determine who’s a real borrower or a fake one? They can use data to analyze patterns of behavior.

Real identities and synthetic identities actually behave very differently according to their data. Real identities typically show up between 18-20 years old and have consistent patterns of behavior throughout their lives. Synthetic identities typically behave in inconsistent and erratic ways, such as moving around a lot or aggressively growing credit. In many cases, sophisticated fraudsters have organized crime schemes in which they produce hundreds of synthetic identities to behave in the same “unusual” way to avoid being flagged for suspicion. While this type of fraud is significantly underreported, it can be prevented with the right fraud detection platform.

2. Stimulus Fraud

Another fast-growing fraud scheme is related to the Coronavirus stimulus. The Small Business Administration(SBA) provided banking services and loan relief efforts to small businesses impacted by the pandemic. Unfortunately, there were potentially billions of dollars improperly distributed due to variations in loan application information. According to the SBA’s Inspector General, the SBA mistakenly approved $14.3 billion in wrong bank accounts, $62.7 billion in multiple loans, and $1.1 billion for ineligible businesses.

Being a clear indication that the SBA had lowered their guard in loan approvals, incidences like this can be further prevented with thorough data analysis. In situations where data is flagged for inconsistencies, the SBA could use case management tools to manage these discrepancies and avoid stimulus fraud. A self-learning data management platform can detect unusual patterns and flag for suspicious behavior. Then, a risk manager for the SBA can immediately look into why these accounts have inconsistent banking information, multiple loan inquiries for the same account, and which ones are actually ineligible for assistance.

3. California’s Unemployment Benefits Fraud

With unemployment continuing to rise across the country, many states are trying to provide additional relief to those without income. Unfortunately, California is now experiencing the largest fraud scheme against taxpayers in state history. The EDD and Bank of America, the issuer of debit cards for unemployment benefits, estimate losses of around $2 billion or more due to fraudulent claims. Many claims were falsely filed under the names of infants, centenarians (people over the age of 100), and people living outside of the U.S.

California could have ultimately prevented these costly fraud losses if they had implemented better software detection tools early on. If they had a data management tool that continually learned from patterns in behavior while also flagging suspicious abnormalities, the EDD and Bank of America could have avoided this record-breaking fraud scheme.

Regardless of each of these banking fraud schemes, there seems to be a relevant pattern that links the three of them together. Banks and credit unions could all benefit from analyzing their data to detect suspicious activity patterns and prevent fraud. Our all-in-one data management platform, NEMESIS, can help banks/government agencies analyze and continuously learn from their data. Suspicious activity can be caught early on and potentially prevent future incidents from occurring. NEMESIS also features integrated case management to make it easier for banks to consolidate fraudulent activity faster and more efficiently.

NEMESIS Solutions

• Speed up the data processing by at least 75% of the time, getting the results faster
• Empower Finance analysts who are not familiar with advanced coding/modeling and transform them into citizen data scientist, focusing more on the actual problems
• Identify fraud schemes with networking and artificial intelligence. The more you use NEMESIS, the smarter it will become and more precise predictions it will make, because it has a feedback mechanism where it learns from its past performance
• Visualize data and provide a supervisor-level dashboard for better management and control
• Assign the case to investigators through the case management system for immediate remedial actions, cutting loss in a min