Effective identification and measurement of model risk for a full view of inventory

Effective identification and measurement of model risk for a full view of inventory

By Tanguy Dehapiot, Head of Valuation Risk, BNP Paribas

Interview ahead of the Model Risk Management Europe Summit
Can you please tell the Risk Insights readers a little bit about yourself, your experiences and what your current professional focus is?

In BNP Paribas, I am the head of the Valuation Risk team, which is part of the Risk Management department. I deal with the normative valuation subject for financial instruments. This relates to valuation control and normative choices for economic, accounting and prudential valuation. This includes the use of valuation models for complex transactions for which there is no observable price.

The main focuses are the prudent value for which a recent regulation has been published, the valuation adjustments and the valuation inputs representing the market information used as arguments of valuation models.

What for you are the benefits of attending a conference like the Model Risk Management Summit and what can attendees expect to learn from your session?

Model Risk is a relatively new subject that started to be discussed in the financial communities in the late 90s. The banking regulators (mostly the US) have published a guidance on Model Risk in 2011, which is now a reference for al supervisory authorities. Banks are more and more challenged on the area of model risk. It is important to share concerns of practitioners in public forums like this Model Risk Management summit where banks and supervisors can share their views outside a conflicting context of regulatory enforcement.

How can financial institutions ensure effective identification and measurement of model risk for a full view of inventory?

Identifying all the possible sources of risk is important. An inventory requires categorisation and classification. This requires setting rules and definition. Although this looks very administrative and legalistic, it is the pre-condition of creating a useable and relevant inventory.

Without giving too much away, could you provide insights on governance of identification?

An effective control requires a phase of review and a phase of approval. Both need to be done by groups that are independent from the Risk taking units. Independent control need to be done with a business mind while being isolated from the business pressure in order to assess the long term economic risk.

Model Risk contain both components of economic risk where judgements and choices must be made (and cannot easily be reduced) and operational risk related to errors that need to be minimised.

What are the implications of new product processes and how can financial institutions overcome this?

A new product process is now a well-established stream of analysis where all risk related to unusual products must be analysed. Model risk is one particular component of this analysis.

How do you see the impact of Model Risk evolving over the next 6-12 months?

Model Risk is now part of the mainstream risk management analysis. It is there to stay but awareness will increase. It should not be seen as an obscure and toxic risk. Bank supervisors have now fully taking it into account. What may evolve is a better management understanding of it and an acknowledgement that it is part of business as usual.

Meet Tanguy Dehapiot at…

Sign up for your free Risk Insights account: