What did we do?

​

We analyzed and tested the cyclical interdependence of a multilateral network of banks. Specifically, we addressed the properties and implications of the unique clearing (payments) vector satisfying the interconnected liabilities between the distinct financial entities (i.e. banks) participating in the clearing network.

 

How did we do it? 

 

First, we developed and compared three different implementations of a static model, namely the optimization algorithm, fixed point algorithm, and fictitious default algorithm. We found that all three methods yielded the same results. Next we developed and explored a dynamic model incorporating discrete time and analyzed how multiple clearing dates affect systemic risk and exposure of the financial system. Finally, we ran the static model using European bank data from 2011 and performed stress tests in which we simulated various financial crises by varying the model’s input parameters. We found that, magnitude dependent, one bank’s default can lead to a greater likelihood of other banks defaulting and have an ensuing rippling effect on the system.

​

Why is this important?

​

This project was worth exploring because the cyclical interdependence of financial entities and the relationship between their obligations to one another is not often considered when analyzing financial risk. However, this consideration is crucial for financial entities because when analyzing their own systemic risk exposure, they must not only consider the financial health of the entities that have direct obligations to them, but also the financial health of the entities to which the issuing entity has linkages. Financial health and credit risk are thus dependent on an intricate liability structure between financial entities.  Failure to consider this dependence can cause tightly connected clearing systems to fail, where a clearing system is a system across which entities make payments to each other and settle obligations. On a broad scale, multiple financial entities failing to satisfy their obligations can lead to a financial crisis with devastating implications on a nation’s economy, such as those caused by the United States (US) crisis in 2007-2008.

​

What's next?

 

In the future, we hope that you and other interested parties can use the models we built to better analyze different realistic scenarios in the financial system and perhaps come up with plans to avoid certain catastrophic economic events.

​

Where can I find more information about this project?

​

We're glad you asked! Click Here to read our entire Capstone Project Final Paper. Our paper also includes snapshots of the specific MATLAB code we wrote for all implementations discussed on this website.