⇽ Back to all course tracks

Margin of Conservatism

Understand the regulation and concepts behind the
Margin of Conservatism for IRB …

Without obligation

Get a free quote

Join others who have saved thousands of hours by getting the right experts.

Content

The Basel concept of a Margin of Conservatism, implemented and further refined in the EBA Guidelines to PD and LGD estimation, manifest the fundamental idea that model risk is credit risk, and should be capitalized for accordingly. The European Capital Requirements Regulation prescribes that institutions should add to its risk parameters estimates a Margin of Conservatism (MoC) which reflects uncertainties in the estimated risk parameters. These uncertainties should be calculated based on a comprehensive process to identify and quantify deficiencies in risk parameter estimation. Biases should be addressed by so-called appropriate adjustments and the MoC should also reflect the uncertainties on these adjustments.

While perhaps conceptually straightforward, the implementation of these regulatory requirements and guidelines is seen as one of the more challenging aspects of implementing IRB. Firstly, because of the complex nature of the requirements and underlying concepts, and secondly, because of the practical difficulties in connecting the theoretical basis and regulation to the practical reality of risk model development and validation.

This advanced course brings to model developers and model validators a thorough understanding of the statistical and conceptual basis of Margin of Conservatism and the applicable regulation, as well as the mental machinery to help bring these into practice. Through challenging hands-on case studies, interactive lectures and stimulating discourse, participants will gain experience with, among others, the following topics

  • Integration of a conceptual and statistical basis of the Margin of Conservatism with regulatory terminology and supervisory expectations
  • Setup, verification and challenge of a process for the identification, quantification and validation of model deficiencies and uncertainties
  • Expected key elements of a Margin of Conservatism framework
  • Approaches for deriving appropriate adjustments and MoC factors for typical sources of bias and uncertainty, such as
    • a lack of data quality, completeness or representativeness
    • limitations to the method of quantification of risk parameters
    • changes to default- and loss definitions, underwriting standards, collection and recovery policies, or business environments
  • Understanding the general estimation error
  • Aggregation of the Margin of Conservatism to the risk parameter level, including Downturn LGD
  • Requirements for the use of human judgment

Learning Goals

After completing the course, the participants will be able to

  • Quantify the Margin of Conservatism for typical sources of bias and uncertainty in credit risk modelling
  • Challenge and verify the completeness, correctness and compliance of a MoC derivation
  • Set up a process and recognize key decisions for the identification of deficiencies and uncertainties in the data and methodology
  • Understand trade-offs between different approaches in terms of complexity, explainability, and conservatism

Target Audience

The course module is intended for:

  • IRB model developers and team leads with a specific interest in determining appropriate adjustments and a Margin of Conservatism for the PD, LGD and EAD/CCF risk parameters.
  • Model validators, Auditors and team leads with similar interests.
  • Supervisors and policy advisors wishing to gain insight in the challenges of regulation with regard to the Margin of Conservatism.
  • Model Owners and Credit Risk Controllers wishing to enhance their understanding of the Margin of Conservatism

Prerequisites

The material will be taught in English. Participants are advised to come equipped with a broad understanding of credit risk, Basel and capital requirements regulation.

  • the calculation of minimum capital requirements
  • modelling the PD, LGD and EAD/CCF risk parameters.

Schedule

The module will be taught over 4 sessions of each 4 hours. The track schedule will be planned in coordination with the client based on the selection of modules.

Format

The material will be taught through active learning and spaced repetition with a focus on acquiring problem-solving strategies for
relevant and realistic case studies.

Example case studies

For this module, examples of case studies are

  • Identify major potential sources of uncertainty, given documentation of a PD estimation methodology
  • Compare the strengths and weaknesses of alternative MoC estimation approaches related to representativeness
  • Explain the relation between model risk and MoC to front office risk management

Please Contact us for details and options.