Latest Basel proposals risk capital arbitrage – highlights of IBM's response to the Basel Fundamental review of the trading book

28 September 2012

The Basel Committee on Banking Supervision invited public comment on its consultative document, ‘Fundamental review of the trading book’. I wanted to share with you highlights from IBM's response, which cautions that the current proposals risk a number of unintended consequences that, in some key areas, could be far worse than the current regulatory framework.

The response, which was submitted by IBM under the Algorithmics name, points out shortcomings of the proposals (supported with real-world examples) and suggests what it believes could be an alternative approach that addresses the Committee's concerns in a more consistent manner.

Ben De Prisco, author of the response and Head of Research and Financial Engineering for risk analytics at IBM, said: “We believe that the draft proposals as outlined by the Committee will create a framework that is rife with the possibility of regulatory capital arbitrage. This framework could lead to situations where a firm’s regulatory capital requirements do not fully reflect their economic risk profile and in certain circumstances understate their economic risk.”


  • Of key concern is the possibility of regulatory capital arbitrage. For example, the proposed approach to capture market liquidity, of using varying liquidity horizons for different positions in the trading book portfolio, may allow firms to select a liquidity horizon based on the impact on regulatory capital savings, rather than an accurate consideration of risks within their economic risk profile.
  • The Committee’s suggestion for using a top-down aggregation approach directly contradicts their 2009 findings, which indicated that such approaches are generally not reliable for aggregating risks.

Alternative approach:

  • IBM Risk Analytics believes the top-down aggregation should be removed.
  • It proposes a ‘Unifying Framework’ instead – which takes an existing approved approach for modelling Counterparty Credit Risk and extends it to include market risk. Many firms already have the infrastructure in place for this.
  • Enforcing a consistent, unifying framework between risk measures reduces the potential for regulatory capital arbitrage as banks would not be able to develop entirely different approaches for each risk measure.
  • From a regulatory standpoint, validation will need to occur only once for both market risk measures and credit risk measures, thereby streamlining and strengthening the supervisory validation process.

Ben De Prisco concludes: “We agree with the Committee that change is needed to fix the ‘mosaic of measures’ in the current regulatory framework, and we recognize that it is challenging to create a framework that can better balance the objectives of regulators and individual firms, with their generally competing desires for prudent capital adequacy and profitable growth. However, we fear that these proposals could result in unintended consequences that are far worse than the current regulatory framework.

“Allowing the same framework for both market and credit risk will provide an even greater incentive for banks to improve their risk methodologies as there is greater re-use of approaches. This consistency will promote real advantages to banks in terms of operational deployment, input data consistency, data mapping, model development and analysis. More consistent and accurate risk insights across the enterprise will lead to better risk-aware decisions, and better decisions lead to better business outcomes for firms."

Become a bobsguide member to access the following

1. Unrestricted access to bobsguide
2. Send a proposal request
3. Insights delivered daily to your inbox
4. Career development