A Dynamic 4s-dsdm Framework for Banking Supervision: Evidence From the South Caucasus Banking Sector Transformation

In the world of finance, banks are the engines that keep economies moving, lending money to build homes, start businesses, and fund daily life. But when these engines run too hot or too cold, the entire system can stall or overheat. For decades, regulators have tried to keep these engines safe using a standard checklist, much like a mechanic inspecting a car. They look at how much money a bank has in reserve, how well it manages its loans, and how much profit it makes. This checklist has worked well when all banks in a region move in the same direction. However, the modern financial world is changing fast. In some places, banks are growing wildly fast, while in others, they are becoming more conservative or shifting their strategies entirely. When every bank is different, a single, rigid checklist fails to see the specific dangers each one faces. The challenge for modern supervisors is to find a way to see these differences clearly and adapt their safety rules in real time, rather than waiting for a crisis to reveal a problem.

A researcher has tackled this problem by studying the banking systems of three neighboring countries in the South Caucasus: Armenia, Azerbaijan, and Georgia. Between 2024 and 2026, these three nations experienced a period of rapid but very different growth. Armenia saw its banks lending money at a record pace, prompting concerns about whether they had enough safety cushions. Georgia's banks grew steadily while reducing their reliance on foreign currencies. Azerbaijan's banks expanded more slowly but with immense strength, leading regulators to introduce new rules to slow down credit growth. Despite these differences, the central banks in each country were still using the same old, static methods to judge risk. The researcher argued that these methods were like trying to measure the weather in three different climates with a single thermometer; they could not capture the unique risks of each nation's banking sector.

To solve this, the researcher built a new, flexible framework called the Dynamic 4S-DSDM. Instead of relying on experts to guess which factors were most important, they let the data speak for itself. They gathered information from every single commercial bank in these three countries—58 banks in total—over a twelve-year period. They fed this massive amount of information into a computer system that used machine learning, a type of artificial intelligence that learns patterns from data. The system analyzed four key areas: how safe the banks were, how stable their funding was, how much profit they made, and how big they were. Crucially, the computer decided for itself which of these areas mattered most at any given time, rather than using fixed rules. It found that the importance of these factors changed as the banks evolved.

Using this smart weighting system, the researcher grouped the banks into three distinct types, or "archetypes," based on how they actually behaved. The first group, called "Strategic Leaders," consisted of banks that balanced safety and profit very well. The second group, "Traditional Savers," were the cautious banks that prioritized stability over rapid growth. The third group, "Aggressive Growth," included banks that were taking big risks to expand quickly. The study found that these groups were not random; they represented real, stable business models that persisted over time. The researcher then tested their findings to make sure they were not just a fluke, using rigorous statistical checks to confirm that these groups remained distinct even when the data was shuffled or viewed from different angles.

The results revealed a fascinating story about how each country's banking system was evolving. In Armenia, the banking sector was showing signs of structural diversity, with more banks moving away from the traditional model toward new, varied strategies. However, the study also found that Armenia's system was highly sensitive to outside shocks. When a major geopolitical event occurred in 2022, Armenian banks reacted quickly and dramatically, shifting en masse toward the safer "Traditional Saver" model. This was a sign of a system that could be fragile under pressure. In contrast, Azerbaijan's banking system was a mix of all three types, showing a high level of diversity that seemed to make it more resilient to shocks. Its banks did not all move in the same direction when trouble started; instead, the different types of banks coexisted, providing a natural buffer. Georgia's system showed a different kind of strength, characterized by "structural elasticity." It had settled into a stable pattern earlier than its neighbors, with a large portion of its banks already operating as "Strategic Leaders," allowing it to absorb stress without collapsing or freezing up.

The researcher also developed a new way to measure how "mature" or stable a banking system is, using a concept called entropy, which in this context measures how mixed or uniform the banks are. A system where all banks are the same has low entropy, while a system with many different types of banks has high entropy. The study showed that while the three countries had different starting points, they all reacted to the 2022 shock in a similar way, with their systems becoming more uniform as banks rushed to safety. This synchronized reaction suggested that regional events affect all three countries, regardless of their individual banking structures.

The most important outcome of this work is a new tool for regulators called the Supervisory Intensity Score. This score tells a central bank exactly how much attention a specific type of bank needs. For the "Strategic Leaders," the score is low, meaning they can be monitored with standard checks. For the "Traditional Savers," the score is moderate, requiring periodic stress tests. But for the "Aggressive Growth" banks, the score is high, signaling that these institutions need constant, targeted audits and stricter rules. This approach allows regulators to stop using a "one-size-fits-all" strategy. Instead of treating every bank the same, they can now apply the right level of supervision to the right bank, based on its actual behavior and risk profile.

The study concludes that the old way of supervising banks is no longer enough for a world where financial systems are changing so quickly. By using data to understand the unique character of each bank and each country, regulators can be more proactive. They can spot a bank that is becoming too risky before it causes a problem, and they can support the banks that are doing the right thing. The research suggests that for the South Caucasus, and potentially for other emerging markets, the future of banking safety lies in flexibility. It is about moving away from rigid rules and toward a system that learns, adapts, and responds to the real, living complexity of the financial world. The findings are not just a theoretical exercise; they offer a practical, evidence-based path forward for keeping the engines of the economy running safely, even when the road gets bumpy.