Dynamic risk management

Date recorded:

Dynamic Risk Management - Outline of proposed DRM accounting model and next steps - Agenda paper 4

Background

In this meeting, the Staff presented the Board with the objectives of the proposed model for dynamic risk management (DRM) and an outline of the proposed model. The focus of the model is to reflect how financial institutions manage the dynamic net interest margin (NIM) resulting from typical banking book assets and liabilities by using derivatives to align the asset profile with the target NIM profile.

Objectives of the proposed model

The Staff have highlighted various shortcomings regarding the current accounting for DRM in previous Board meetings. These shortcomings can be summarised as follows:

1. Lack of transparency on management’s target profile

Transparency on management’s target profile and its implications for future cash flows are largely absent from financial reporting today. This makes it difficult for a user to understand and evaluate how management manages interest re-pricing risks and the rationale for their decisions. It also limits users’ ability to compare management’s approach to DRM with other companies and across time.

2. The capacity issue

This refers to an entity not having enough hedged items against which derivatives could be designated in a hedging relationship. Typically, an entity needs to use two interest rate swaps to align a particular asset profile with the target profile – one to cancel out the existing profile and the other to create the desired profile. The derivative used to cancel out the existing profile would hedge the cash flows from the loan asset. The derivative that is used to create the desired profile is supposed to hedge the interest rate risks arising from the demand deposits (assuming this is the source of funding). However, demand deposits are generally regarded as zero-rate perpetual funding that is not subject to either cash flow or fair value interest rate risks (see AP 4 to the September 2017 Board meeting for detail). Consequently, such demand deposits are not an eligible hedging item, giving rise to the capacity issue. To circumvent this issue, entities currently try to find alternative eligible items on the balance sheet so that hedge accounting can be applied. However, this does not faithfully represent the entity’s risk management activities.

3. Operational challenges arising from the dynamic nature of DRM

IFRS 9 requires a one-to-one designation between hedged items and hedging instruments. However, by definition, the assets and liabilities managed under DRM are constantly changing – existing loans mature and new loans are originated all the time. This requires management to update their derivatives portfolio constantly to align the new asset profile with the target profile, which results in frequent designation, de-designation and rebalancing of hedging relationships. Such frequent adjustments are complex, costly and prone to operational error.

4. Inability to evaluate management’s performance properly

Current financial reporting does not indicate whether management was successful in achieving the target NIM profile. While current hedge accounting provides some information about the effectiveness of the hedge, the results are marred by the capacity issue and the use of ‘surrogate’ hedged items to complete the hedging relationship.

Each of these shortcomings translates into an objective of the proposed model. In other words, the proposed model would consider hedge designations on a portfolio basis to address the capacity issue, thereby reducing the operational challenges currently faced by entities. The proposed model further aims to increase transparency on management’s target profile and to create a performance metric that would demonstrate whether management was successful in achieving the target profile through appropriate disclosures.

Outline of proposed model

The Staff considered accounting for DRM using the mechanics of either a cash flow hedge or a fair value hedge. On balance, the Staff believed that the cash flow hedge mechanics is more suitable because the objective of DRM is to manage cash flows rather than to eliminate fair value risk. The Staff acknowledged that there are shortcomings associated with the cash flow hedge mechanics, not least the rationale for the ‘lower of test’ and deferring changes in fair value of the hedging instrument in other comprehensive income and their subsequent recycling to profit or loss. However, the Staff believed that there are more conceptual challenges associated with the fair value hedge accounting mechanisms when applied to DRM, most notably because recognising the target profile’s change in fair value on the balance sheet may not meet the definition of an asset or a liability.

Next step

The Staff will discuss the following aspects of the proposed model in future meetings: (a) asset profile, (b) target profile, (c) DRM derivative instruments; (d) effectiveness assessment, (e) criteria for designating a hedging relationship, and (f) disclosures.

Staff recommendation

The Staff recommended that the Board adopt the cash flow hedge accounting model for accounting for DRM.

Discussion

The Board generally supported adopting the cash flow hedge accounting model for accounting for DRM.

The main point raised by the Board was the need to have proper project management. The Staff should set project milestones so that they can test the model with stakeholders and have early feedback on its feasibility before the Board invests more time and resources in developing the model. Many Board members cautioned against spending years in developing a model that no one will use if it does not suit the stakeholders’ needs. They noted that early identification and discussion of the challenging areas will be crucial and that seeking feedback from users and prudential regulators first could help assess whether the model provides useful information as this could help reduce the ‘noise’ from reluctant preparers. This is especially the case for prudential regulators as applying cash flow hedge accounting would result in more volatility in equity. There is also a need to think beyond European banks and consider the global population when developing the model.

Some Board members also noted that in theory it is good to require disclosure of the target profile, but in practice banks may not be willing to disclose it because it is proprietary information and many inputs are considered commercially sensitive. This will have a knock-on effect on the ability to assess management performance. These practical difficulties will all affect the ultimate success of the model.

Correction list for hyphenation

These words serve as exceptions. Once entered, they are only hyphenated at the specified hyphenation points. Each word should be on a separate line.