4 Ways to Detect Fraudulent Transactions Using Behavioral Economics

4-trends-behavioral-economics-spot-fraudulent-transactions.jpg

GUEST BLOGGER
Debanjan Chatterjee

The design of effective anti-fraud solutions requires a clear understanding of fraudsters’ psychology. Additionally, a strong grasp of cognitive, emotional, cultural and social dimensions goes a long way in deciphering the modus operandi of criminals.

This is where behavioral economics comes in. If you’re not familiar, behavioral economics studies the decision-making processes of individuals and how their choices differ due to various nuances of emotional states, cognitive ability and personality.

In the context of fraud risk, such subtleties have the potential to illuminate a wide variety of crime trends, which in turn, can aid with your organization’s design of anti-fraud solutions. Here are four ideas from behavioral economics that can help you spot fraudulent transactions.

1. Loss aversion

Studies have shown that our response to loss is stronger than our reaction to a similar gain, meaning that people by nature are averse to damages. In the study titled “Advances in prospect theory: Cumulative representation of certainty,” the authors analyze how anticipated disappointment influences a person’s actions.

Example

A growing portion of credit card fraud can be traced back to marketplaces in the dark web, where cybercriminals sell stolen card information. Fraudsters buy such data to try purchasing goods and pushing the stolen cards to their credit limit. Consequently, declined card transactions highlight the prospect of losing the money a fraudster spent in acquiring such data. Anticipation of loss pushes criminals to try out the stolen cards repeatedly in the dire hope of clawing back their investment. Such a response to card declines radically differentiates from genuine consumers, who might try once more but would then use another card in such circumstances. Tracking and flagging this behavior in card transaction patterns can be used to detect fraudulent activity.

2. Anchoring effect 

This is a cognitive bias whereby an individual’s decisions are influenced by a particular reference point, or an “anchor.” In a non-fraudulent example, a company may place expensive products alongside cheaper ones, which influences consumers to use the higher priced product as the anchor. This way, any price lower than the anchor price seems attractive, irrespective of the actual worth of the lower priced product.

Example

When examining fraudulent transaction activity, we are likely to witness subsequent attempted transactions that gradually climb in ticket size as fraudsters use the first approved transaction’s amount as the anchor. Similarly, for declined transactions, a gradual downward trend is observed, as criminals attempt to slip just below the threshold amount at which a bank’s fraud detection system starts raising alarms. Such patterns in successive ticket sizes differ from genuine card usage, where the transaction amounts do not strictly increase or decrease over time, but vary according to the need of the cardholder.

Anchoring bias also highlights the importance of successfully identifying fraud from the very first transaction itself, to deter criminals from attempting high-value activity.

On an interesting note, multiple studies have shown that individuals most likely to demonstrate this bias are usually a) introverts, b) have low expertise and cognitive ability and c) exhibit overconfidence.

Cognitive ability can be expected to be high for fraudsters operating as a part of a syndicated fraud ring. Thus, spotting such patterns in data can be used to sketch their holistic profiles.

3. Psychological profiling

Studies in behavioral economics contain a wide variety of experiments that can be used to understand the psychology of consumers. Accordingly, such insights can be leveraged by fraud detection systems to raise alarms if patterns in product usage do not match the known cognitive profile of the customer.

Example

Some potential fraud alerts might be:

  • A customer who has been observed to be risk-averse uses their credit card for a large purchase at a cryptocurrency exchange

  • A cardholder who likes a routine way of life buys groceries at an erratic hour.

  • A financially prudent customer with a steady focus on savings splurges on luxury goods.

4. Authority bias and cognitive overload

Soon after the onset of COVID-19, there has been a dramatic rise in scams and theft of sensitive information via social engineering and phishing. Criminals have been duping people by impersonating health officials and vaccine providers. Vulnerability to such fraud vectors depends on following factors:

  • Authority bias: The tendency to be more influenced by people who appear to be experts. This makes individuals trust impersonators of persons in positions of authority, such as people who pose as bank managers, doctors and investment gurus.

  • Cognitive overload: Scammers often create illusions of fear and urgency. This tactic places undue stress on mental abilities, leading individuals to behave irrationally and fall prey to fraud schemes.

Example

Organizations may want to implement special efforts to identify consumers who are especially vulnerable to the above influences in order to design a targeted customer education program.

What to do now

The modus operandi of fraudsters is known to morph at a feverish pace. Emerging technology, changing geopolitical equations, newer financial products and shifts in the legal landscape all play a crucial role in shaping criminal incentives. In this increasingly complex environment, tools such as behavioral economics can go a long way in helping fraud fighters glean insights into the nuances of fraud propagation.

Debanjan Chatterjee is a fraud risk management professional and has spent more than 13 years designing anti-fraud solutions.


SOURCE: ACFE Insights – A Publication of the Association of Certified Fraud Examiners