Detecting Ponzi and other shady schemes
Since money has existed, there have been people devising ways to defraud others. A Ponzi scheme is one of the oldest forms of fraud, and continues to be rampantly active today.
Built on a simple concept, Ponzi schemes are financial crimes that involve an unscrupulous person exploiting individuals by promising them a high return on their investment after a fixed amount of time. Naturally the return is much higher than a regular broker could offer, and returns increase with a longer commitment period. Any funds withdrawn early face a significant penalty.
As the saying goes, the person running the Ponzi scheme then ‘robs Peter to pay Paul’, using the funds of newer investors to pay phony returns to earlier investors. Meanwhile, the con artist is most often using the funds to finance a lifestyle of luxury. Eventually the money runs out, the scheme collapses, and the ‘investors’ have lost their money.
The Ponzi challenge
While there are certainly ways for individuals to determine if they have been targeted by a Ponzi scheme (including your financial advisor: offering overly high returns with little to no risk; taking custody of the money; providing consistent returns regardless of overall market conditions; lacking investments controlled by securities regulators; and building a client base through almost solely through referrals), it’s significantly more challenging for financial institutions to detect a Ponzi scheme.
It can feel a lot like trying to find a needle in a haystack of accounts and transactions, and requires information that many organisations don’t have access to, including big picture information and the ability to track large volumes of transactions. It becomes more challenging given that information about transactions and actors is not publicly available to proceed to a verification or control.
Finding the fraud
When it comes to detecting a Ponzi scheme, businesses must know the crime’s red flags in terms of accounts and transactions. There are several ways to detect and identify these key indicators of a Ponzi scheme:
An important way to determine if a bank account is being used for a Ponzi scheme is to profile the connections and transactions involved in a given account. By analyzing the densification of connections in a network – that is, how many people are connected to the account – over a specific period of time, you can detect trends.
With a Ponzi scheme, there will typically be growth over time with more people investing in the fake actions. A top performing financial advisor, for example, would typically add five to ten new clients to an account each month. Numbers higher than that can be indicative of fraud.
In addition, network analysis can be used to follow the path the money has taken to help determine if it has truly been invested elsewhere or if the account holder is actually ‘stealing from Peter to pay Paul’.
Monitor for repetitive transactions
As seen in the infamous Bernie Madoff Ponzi scheme, Madoff frequently made repeated transactions to and from the same individuals, seemingly without reason. In one case he transferred the same amount of money—almost one million dollars—to one customer 318 times, sometimes even on the same day.
These repetitive and high-dollar value transactions often did not have any impact on the net worth of the account, with the total monthly amounts going in usually equal to the amount going out. While the purpose of these transactions isn’t always clear, they are certainly suspicious and may be indicative of a Ponzi scheme or other type of fraud.
To facilitate effective network analysis and the monitoring of suspicious and/or repetitive transactions, organizations will increasingly need to leverage machine learning in addition to a rules-based approach. By providing the machine with historical data on transactions conducted by fraudsters, the machine can be trained to learn that certain types of transactions are related to fraud.
After the machine is trained, it will observe current transactions and issue an alert if a transaction looks similar to the pattern it has been trained to detect. By training the machine to learn what type of transaction has historically been related to a Ponzi scheme, organizations can actually prevent the fraud from taking place, resulting in highly reduced financial loss and exposure.
New technologies, new detection methods
With 59 Ponzi schemes uncovered in the U.S. in 2016 alone, it’s an investment fraud scheme that won’t be disappearing anytime soon. Although regulators have increased efforts to reveal these schemes following the Madoff case, organizations must continue to take steps to protect themselves from the reputational and legal ramifications they can face for being connected to a fraud scheme. Technology and approaches such as network analysis and machine learning will be key in this battle.
About Andrew Simpson
Andrew Simpson (LinkedIn | Twitter) is Chief Operating Officer at CaseWare RCM and has more than 20 years of experience building businesses in the fields of information systems audit and security, data analytics, Anti-Money Laundering and forensics. He is a regular contributor to conferences and a recognized thought leader in financial crime management.