IRS 2019 MMI-SMA-BC

The Interbank Money Market (MMI) is a complex market that is essential to the proper functioning of a country’s banking system and economy. The primary objective of this project is to design and develop secure and agile tools and organizational structures that can bring greater stability and resilience to the MMI system.

Two research questions have been identified: What learning methods and evolving forms of knowledge help maintain trust among market participants? What market structures and participant strategies are most effective in stabilizing interbank market rates?

To address these questions, we propose a mixed-methods research approach based on an exploratory qualitative method, supplemented by a quantitative method involving flow simulation. This calls for a multidisciplinary approach, bringing together researchers in management sciences (finance and information systems) and computer science. Similarly, a future doctoral student with expertise in both finance and computer science will be recruited.

The complexity and dynamics of the MMI system are managed using a decentralized, self-organizing approach based on a multi-agent system (MAS). Autonomous agents represent the various entities within the system, and interactions are managed through negotiation protocols between intelligent agents that adhere to MMI rules. The recording of securities sales and purchases takes place in a central securities depository system based on blockchain technology, providing the MMI with greater transparency, security, traceability, speed, and cost savings. An IT platform will be developed for the simulation, combining both the multi-agent system (MAS) and blockchain technologies. The data for the simulation of interbank transactions will be based on real-world data provided by the Crédit Agricole Group and the Banque de France.

Various MMI scenarios will be tested under symmetric shock conditions (all banks are affected in the same way) or asymmetric shock conditions (one bank or a certain type of bank is affected) in order to assess the system’s stability and resilience.

Collaborators: Paul Reaidy (project leader, Assistant Professor with HDR, CERAG); Philippe Madiès (Full Professor, CERAG); Julie Dugdale (Assistant Professor with HDR, LIG); Morteza Allaeddini (Ph.D. candidate, CERAG)