Although geopolitical risk is not a new phenomenon, it has emerged as one of the most critical risks in the payments industry today. Recently, Jamie Dimon, CEO of J.P. Morgan, referred to geopolitical risk as “the thing that most concerns me”. 

Financial institutions and payment companies are not new to the adverse impacts of geopolitical risk. Sanctions and the tools used to enforce them, for example, are but one of the well-established tools used in financial services relating to such risks. 

However, the evolution of sanctions and the increasingly complicated manifestation of those sanctions highlight that geopolitical risks are often hidden within our economies. The rules-based approach and manual investigation-based systems that worked in the past are no longer sufficient for identifying and preventing the manifestation of such risks.

In the ever-evolving global political and economic landscape, the intersection of data science and geopolitical risk management has become critical for cross-border payments. 

The nuanced and indirect nature of geopolitical and sanctions risk – sanctions circumvention is widely viewed as a critical risk at the moment by both regulators and financial institutions – together with customer and regulatory expectations for ever-faster payment execution, creates the potential for a new level of risk. As has been the case for the past 30 years or more, financial institutions will be expected to independently mitigate that risk or be responsible for the consequences.

So geopolitical risk and financial crime risk, including both money laundering and sanctions, are intertwined. This much has been clear to the market for as long as country risk has been assessed. What has changed is the complexity and opacity of payments, combined with a level of velocity which amplifies risk. 

For that reason, the tools which evolved historically to mitigate these risks are no longer fit for purpose. Real-time payment analysis will increasingly become key to identifying and managing geopolitical, financial crime, reputational and regulatory risk. 

A case study on payment risks

As with so many areas of risk today, data science is a critical component to solving this problem in an effective and sustainable way. Recent research conducted by the team at Elucidate highlights the potential application of data science in navigating the complexities of cross-border financial flows, particularly in the context of sanctions.

Elucidate presented our findings at the 2024 Empirical Anti-Money Laundering Conference organised by the Central Bank of the Bahamas, examining the web of cross-border transactions involving jurisdictions of the Commonwealth of Independent States (CIS). 

The paper suggests that “from a risk management perspective the FATF and EU lists are of limited practical value in identifying jurisdictional-level sources of risk. A more empirically-driven approach to cross-border ML/TF risk at the jurisdictional level would contribute to more appropriately targeted focus of limited regulatory and compliance resources, improving overall systemic effectiveness”. 

By harnessing a cross-institutional dataset of transactions between 2018 and 2023, we applied a structured risk model analytical framework that integrates trade flows, GDP, geopositioning analysis, graphic distance and indicators of money laundering risk.

This revealed transactions representing a high probability of association with sanctions circumvention, none of which would have been highlighted by conventional sanctions detection software.

Our findings reveal that despite various CIS jurisdictions’ efforts to enforce compliance with international sanctions, certain cross-border flows the widely reported risk of sanctions circumvention is real and is manifesting in market activities. 

Furthermore, the data relating to these payments are not necessarily correlated to standard high-risk country metrics, such as those published by the Basis Institute and others. This further underscores the limitations of current AML frameworks and the necessity for a more nuanced, data-driven approach.

These findings, which also examined the use of traditional financial crime typologies as a means to backtest and determine exposure to evolving geopolitical and sanctions risks, revealed decidedly mixed results. 

Whilst there were occasions where typologies familiar to transaction monitoring teams – round dollar amounts, offshore customers, etc. – were associated with sanctions circumvention, it was rare that any single typology was uniquely present. More often a combination of typologies existed. 

Complex risks require an updated tech stack

Such complexities would render transaction monitoring tools quite ineffective at highlighting the most acute risks. This highlights the critical need for a detailed analysis of transaction patterns that may or may not appear to diverge from the norm. This also requires a data science-driven approach to uncovering potential financial crimes, as the volume of data and complexity is inaccessible to even the most seasoned investigator.

We urge companies and regulatory bodies alike to recognise the criticality of data-driven insights as a core element in financial security and compliance. This is not the automation of risk management, but rather the augmentation of existing control frameworks with data science to ensure they remain fit for purpose.

By embracing the methodologies exemplified in our research, stakeholders can develop more effective risk management frameworks, tailored to the unique and evolving geopolitical risks and challenges in which they operate.

More work is needed to integrate these new approaches into the payments landscape, but it is essential to ensuring both regulatory compliance and the overall integrity of payments platforms. Real-time solutions and solid risk modelling are essential parts of a robust payments tech stack. 

Shifting the control framework left, with all the benefits that data science, machine learning and generative AI offer, will ensure that we create machine speed compliance capable of matching the increasing velocity of the payments ecosystem.

Not all financial institutions will be able to do this alone, but fortunately, there are tech partners like Elucidate that enable change and evolution to keep pace with even the most fast-changing environments.