Building a CCAR compliant operational risk stress model

Building a CCAR compliant operational risk stress model

We interviewed Christiane Hoppe-Oehl at UBS to gain insight on her professional experience as Head of Operational Risk and Reverse Stress Testing. Christiane will be presenting at the Stress Testing Europe Summit taking place in October 5-6.


Christiane, can you please tell the center for financial professionals’ readers about your self and your professional experience?
I hold a diploma in mathematics. I have worked 20 years in the financial industry covering market, credit and operational risks. I covered market risk at Bayern LB in Munich and Paris. I was responsible for market risk modelling, trader supervision, the development of pricing tools, as well as trading, pricing, hedging and structuring of Structured Equity Products. I joined UBS 10 years ago in credit risk modelling where I put in place a stress governance enhancing the banks risk frameworks. Since two years I am responsible for the operational risk stress model and enhanced it to comply with CCAR standards. Apart from operational risk, I am responsible for reverse stress testing and other consequential risks under stress.


We are looking forward to you presenting at our upcoming Stress Testing Europe Summit where you will review the building of CCAR compliant operational risk stress model. What are the advantages of using quantitative models over an expert judgment?

Not long ago, US regulators preferred quantitative models but the trend is moving towards more expert involvement to make up for data issues and enable estimates of future vulnerabilities which might differ substantially from historical observed losses. However, regulators still want banks to analyze their past data and get a good understanding of relationships to macro-economic events instead of purely relying on experts who might be biased or might have very subjective views on potential losses. Losses observed in operational risk were substantial for many banks over the past 10 years. Increased regulatory scrutiny has led to ever higher fines amounting to billions of losses per bank. There is significant model risk involved and results might differ substantially depending on the approach taken. Key is a thorough sensitivity and benchmarking analysis to better grasp the correct range of possible future operational risk losses and the resulting impact on the banks capital.


Without giving too much away, how do you make a CCAR compliant operational risk stress testing model forward looking?

It starts with a diligent risk identification process, involvement of business, senior management and various internal parties to ensure the current most material risks can be identified and measured. This process helps to raise awareness about key risks and improvements in risk management. Dwelling deeply into what could go wrong and assessing the possible loss amount will lead to the right measures being taken before the risks will materialize. Understanding and therewith preventing events which could hurt the bank most is at the core of risk managements role.


What are some of the challenges in estimating legal losses under stress?

Legal losses have increased substantially over the past years and the trend for high fines hasn’t reversed yet. It is hard to predict possible outcomes. Close collaboration between Legal, Risk, and Finance is required to be able to raise awareness and estimate capital impacts while confidentiality is key. Large legal fines can have a significant impact on the companies’ stock price, clients and employees, reputation and the brand.


How do you see the role of the Stress Testing Professional changing over the next 6-12 months?

Modelers must learn to sell their models and educate the business, to move away from pure quantitative assessments to a more process oriented role. The business must learn to judge on models and understand their impact on capital. Collaboration between different entities (Risk, Treasury, Finance, Legal) must improve to get a conclusive bank-wide picture.