The mounting challenge of trade for AML compliance
Trade financing is a big business for banks, with letters of credit and an important revenue stream. For the developing world, global trade is key to economic growth.
The global trade finance market, with an estimated value of $5.2 trillion, facilitates the cross-border movement of goods and services around the world. It includes buyers and suppliers of all sizes, with financial institutions providing the liquidity and the risk assessment necessary to execute trade transactions.
At the same time, the Financial Action Task Force (FATF), a global financial crime watchdog, has identified trade as one of most worrisome methods by which criminal organizations and terrorist financiers move large amounts of money while disguising origins and integrating the funds into financial systems.
A growing global problem in the past several years, international trade attracts bad actors and terrorists aiming to exploit legitimate types of activities while manipulating and misrepresenting the quality, value, and or quantity of imports and exports.
Indeed, trade-based money laundering (TBML) is one of the most sophisticated methods of “cleaning” dirty money, notably for narcotics traffickers and terror organizations. Due to the complexity of trade finance and global shipping logistics, money-laundering red flags are usually manyfold and also among the hardest to detect. While the scope of money laundered through the international trade system is also unknown, as estimates range vastly from hundreds of billions to trillions of dollars per year.
As a result, many banks often choose to de-risk or limit activity, meaning businesses and economies lose access to global trade.
To help address the challenges of detecting trade-based money laundering, the FATF has asked financial institutions to redouble efforts and reassess controls to ensure they are using the right tools in combatting TBML. In March 2021, a new guidance issued specific guidelines for trade-based money-laundering risk factors.
Adopting a risk-based approach
Classic money-laundering flags used in a rules-based approach, such as threshold values and high-risk jurisdictions, are simply too general for the complexity of global trade where so many factors are involved in the transaction.
The FATF highlighted a host of risk indicators that can point to TBML in the guidance. The multitude of risks is grouped as follows:
Structural risk indicators such as high-risk jurisdictions, suspected use of shell companies, a lack of public information or corporate offices, little or no business activity, and criminal records.
Trade activity risk indicators include a discrepancy between the line of business and the trade activity, use of many intermediaries, abnormal shipping routes, unconventional use of letters of credit, importing wholesale commodities at or above retail value unreasonably low-profit margins, a newly formed or recently re-activated trade entity engage in high-volume and high-value trade activity.
Trade document and commodity risk indicators for example inconsistencies across contracts, invoices, or other trade documents or prices that do not seem to be in line with commercial considerations. For example, when the value of registered imports of an entity displays significant mismatches to the entity’s volume of foreign bank transfers for imports.
Account and transaction activity risk indicators such as cases when an account displays an unexpectedly high number or value of transactions that are inconsistent with the stated business activity of the client transfers to a trade-related account are split and forwarded to non-related multiple accounts, cash deposits are just below reporting thresholds, or payments are routed in a circle.
Furthermore, the FATF notes that a single indicator may not on its own warrant suspicion of TBML or a clear picture of the activity, as some risk indicators require the cross-comparison of data elements.
The power of AI-driven analysis
AI-based solutions are experiencing accelerated adoption by financial institutions in search of more sophisticated solutions to manage compliance operations. Trade finance is a key area of financial activity with the need to connect the dots using machine learning and AI for AML programs.
To make risk-based analysis, the most effective method available today is artificial intelligence deploying advanced machine learning that can detect and isolate abnormal cases from routine trade activity within complex sets of data. AI can effectively calculate the risks to indicate suspicious activities outside of expected behavior in commerce.
This kind of system can analyze a multitude of risk factors and pinpoint abnormal cases. Data such as SWIFT trade codes such as MT 103, 202, and the 400, 700, and 900 series codes, KYC, country risk factors, unusual or out of pattern shipment origin, industry codes, letter of credit amendments, and many to one counterparty, unusual in-out ratios, a suspicious number of changes that were added to a contract with the same incoming and outgoing payment amounts, like a pipe account.
Allowing the data to lead to abnormalities, instead of telling the system what to look for in a rules-only based method, allows AI to detect new and unpredictable typologies or “unknown-unknowns” in trade.