May 2012

Trading book regime under scrutiny…

Basel Committee consults on the fundamental review of trading book capital requirements

On 3 May 2012, the Basel Committee on Banking Supervision (BCBS) released its proposals for a ‘Fundamental Review of the Trading Book.’  The upheavals of the financial crisis highlighted a number of failures in the trading book regime – including inadequate capital held against market risk and wide latitude in determining which assets could be placed in the trading book. The BCBS seeks to address these issues with suggestions for wide ranging change to the identification and risk measurement of trading book assets. Its objective is to build on other changes to the market risk capital requirements with a trading book regime that promotes comparable capital outcomes between banks. The proposals may result in significant change to existing models, higher capital requirements – particularly for credit instruments and some derivatives - and substantial increases to the resources needed to develop and monitor trading book risks.

In summary, the BCBS proposes:

Changing the underlying methodology from ‘value at risk’ (VaR) to ‘expected shortfall’ (ES) - a fundamental change to existing regulatory capital models. Such a move would add modelling complexity, and increase capital requirements for many assets.

Changing the basis of definition for trading book assets – either by significantly enhancing the operational requirements to support an ‘intent to trade’ to justify a trading book designation or by moving designating trading book assets based on whether or not they are capable of being fair valued. In either case, an ability to re-designate an asset once designated to either the trading or banking book will be much more limited.

A more detailed assessment of illiquidity risk - with additional capital add ons for instruments at greater risk of illiquidity under stress.

Harmonisation of differences between internal models and the standardised approach - This could include closer alignment of allowable approaches to hedging and diversification, limiting the potential capital benefits of internal models relative to the standardised method, and significantly enhancing the operational requirements around the approval and maintenance of internal models.

Impact on firms

The overall effect of these proposals, if agreed, would be potentially significant reductions in the benefit available to users of internal models, given increased operational costs and restrictions on capital benefits. More fundamentally, changing modelling approaches to ES would require a major commitment of time and resources from banks, and would also be likely to drive further increases to capital requirements. The complexity of ES - along with changes to regulatory requirements for liquidity and operational requirements to validate the effectiveness of models - will not just impact the level of capital. Auditors face increased challenges to understand, audit and test capital charges. Similarly, Boards and senior management must consider how they can build their own capabilities to assess and act upon ES model outputs when making risk management decisions.

Our view

Most industry participants agree that some change was necessary. Wide discrepancies between outputs of internal models – highlighted in the UK Financial Services Authority’s (FSA) own analysis – support the view that standards around internal models need tightening. Similarly, enhanced alignment and transparency of outcomes between internal models versus the standardised method will aid regulators seeking to look at aggregate risks and spot potential issues with model based approaches.

But these proposals come at a time when banks are already diverting significant time and money towards regulatory change. New capital charges under Basel 2.5 and 3, requirements for central clearing of derivatives, market and regulatory driven increases in collateral – these regulatory developments and more will, together, fundamentally change the dynamics and economics of trading. The cumulative effect of these changes has not yet been fully assessed, with potential repercussions not only for banks but also liquidity and efficiency in the wider financial market. These proposals, and their interaction both with one another and with other ongoing regulatory change, should therefore be carefully assessed by banks.

Comments on the proposals are due by 7 September 2012.

In the detail…

An overview of the key proposals set out by the BCBS and our view of the potential implications are set out below.

1. Move from using Value at Risk to calculate regulatory capital to models based on Expected Shortfall (ES). Both internal models and the standardised approach would migrate to using an ES methodology.
1. ES capital charges capture potential losses above a set confidence level – unlike VaR. The result is a potential increase in capital requirements for assets more likely to ‘jump’ from lower risk to higher risk of prospective loss at the extreme ‘tail’ of probabilities. As a result, some asset classes such as structured credit and complex derivatives may become uncompetitive.
2. Banks have invested significantly in VaR models and more recently stressed VaR capabilities. Moving to ES would require supplementing or replacing these models – with r major investment of time, resource and money.
3. ES capital charges incorporate up front a potentially large loss under unlikely scenarios. Actual daily losses under normal trading conditions will appear relatively small when compared with ES potential losses. ES based models may therefore be less useful in alerting management to developments in market conditions unless they have a clear understanding of the ES approach and how to monitor against it effectively.
4. ES calculations are also more complex – making it more difficult for senior management to understand ES based risk reports and use them effectively for their own scrutiny of risks.
5. ES methods require large amounts of data to predict potential losses, but cannot fully predict what has not yet happened. Over reliance on complicated models was an issue in the previous crisis which is not necessarily addressed by the move to ES.
2. Two options to change the methodology for defining trading book instruments – ‘trading evidence’ or ‘valuation’:
   
1. ‘Trading evidence’ extends the concept of having an ‘intent’ to trade with the ability to demonstrate capability to trade and risk manage the asset – including daily mark to market at fair value. Additionally, there would be much stricter limits over movement of assets out of the trading book after their initial classification.
2. ‘Valuation based’ moves away from ‘trading intent’ and instead by using accounting ‘fair value’ rules to determine which assets are allocated to the trading book.
Either option will significantly reduce the flexibility banks have had in practice to move instruments between books in order to apply a more favourable regulatory treatment.  However, given new market risk capital requirements under Basel 2.5 and 3, the previous advantages are already significantly eroded.  Therefore while either proposal may drive significant changes in bank processes, the impact on regulatory capital may be more muted.


3. A new three stage approach to illiquidity:
1. Introduces five brackets of market liquidity into which firms must allocate all assets. The brackets reflect the expected time to exit or hedge the asset in stressed market conditions, from one day to one year.
2. Applying a capital charge to the risk reflected in each bucket (presumably as a percentage of the assets in each bucket)
3. Additional capital add-ons for assets at higher risk of ‘jumps’ in liquidity premia (examples could include Collateralised Debt Obligations (CDOs) and bank Floating Rate Notes (FRNs)) – identified relative to criteria to be defined by the BCBS
Capital requirements will almost certainly rise for some instruments. New charges for liquidity, combined with likely higher capital charges under ES, may significantly reduce the profitability of many credit based businesses.


4. Changes to treatment of hedging and diversification – reducing the risk reducing benefits under internal models, but potentially enhancing these under the standardised model.
These changes, along with other limits and operational changes for internal models described below, may significantly narrow the relative attractiveness of model based methods versus the standardised approach.


5. Harmonising outcomes between internal models and standardised models by:
   
1. Establishing a closer link between the two approaches – including greater harmonisation of approaches to hedging and diversification (as above)
2. Requiring mandatory calculation of the standardised approach by all banks
3. Considering the merits of introducing a floor or surcharge (based on capital requirements under the standardised method) which would place an upper limit on the benefits (ie reduction in capital) available to users of the internal models based approach
The reporting of a standardised approach would make it easier to compare and contrast capital calculations between firms, and many banks to some degree already shadow internal model outputs with standard method calculations to validate performance of the models. It is likely that capital will increase for banks currently using internal models – but there may be some decreases for instruments valued on the standardised method when additional hedging and diversification benefits are realised.


6. In addition, the BCBS proposes strengthening the rigour with which internal models are both approved and then monitored for effective performance.  This would be achieved by:
1. Breaking the model approval process into smaller, more discrete steps which would allow individual desks to be exempted from using internal models
2. Two quantitative tools to measure the performance of models: desk based P&L attribution, which links daily movements to risk factors in regulatory models, and daily back testing of actual to model forecast losses by individual trading desk.
3. Banks whose models are deemed inadequate at desk level (or more widely) would have model approvals for those activities withdrawn.
These new requirements could drive significant operational challenges for many banks.  While both P&L attribution and back testing are a core part of current financial and risk management frameworks, these exercises - as well as regulatory capital models - are often done at portfolio level. Capability to model and monitor for regulatory purposes by desk for many banks will mean employing significant additional resources and management time.


7. Revisions to the standardised approach, intended to emphasise ‘simplicity, transparency, consistency and improved risk sensitivity.’  Two alternative approaches are considered:
1. Partial risk factor approach – groups instruments into approximately twenty ‘risk buckets’ across five asset classes, with better attribution of hedging and diversification benefits, and inclusion of individual instruments in multiple buckets where exposed to multiple risks
2. Fuller risk factor approach – Maps instruments to prescribed regulatory risk factors subject to stress testing and some improved recognition of hedge benefits.  The aggregate risk position would be subject to a simplified regulatory algorithm to calculate the capital charge.
Excluded at this stage is the treatment of interest rate risk in the banking book, where further work is expected later in 2012, and greater integration of market and counterparty risk.

Back to top

Unsubscribe | Privacy | Legal

© 2012 KPMG International Cooperative (“KPMG International”), a Swiss entity. Member firms of the KPMG network of independent firms are affiliated with KPMG International. KPMG International provides no client services. No member firm has any authority to obligate or bind KPMG International or any other member firm vis-à-vis third parties, nor does KPMG International have any such authority to obligate or bind any member firm. All rights reserved.

KPMG International, 15 Canada Square, London, E14 5GL

Designed and produced by RR Donnelley.
Publication Number: RRD-270151