Revised Guide to Internal Models (EGIM)

The European Central Bank (ECB) has revised its Guide to Internal Models (EGIM) to ensure consistent application of supervisory standards and reinforce the interpretation of European Union (EU) rules on the use of internal models. The revised document consolidates the experience gained in supervision and seeks to harmonize practices among institutions, integrating the most recent regulatory changes.

Revised Guide to Internal Models (EGIM)

Watch video

 Executive summary

The revised Internal Model Guidance reinforces the fundamental principles of model governance, with increased requirements for documentation, data governance and consistency at the consolidated level. It introduces new sections on climate and environmental risks, the use of machine learning techniques and accountability in consolidation and outsourcing processes. The guide also outlines enhanced expectations for internal credit, market and counterparty models, aligned with the latest regulations such as the Capital Requirements Regulation (CRR) III.

The updated version of the Guide will be used as a reference for Supervisory Review and Evaluation Process (SREP) assessments from 2025 onwards.

Main content

This Technical Note summarizes the highlights of the EGIM revision:

• General principles. The requirements for model governance, documentation, data governance, validation and internal audit are consolidated. Specific guidance is added for environmental and climate risks, the application of modified or extended models, and third-party involvement. In addition, clear expectations for the use of machine learning are defined, including criteria for explainability, justified complexity and data standards.
• Credit risk. Criteria for the deployment and permanent partial use of the internal ratings-based (IRB) approach are clarified, references to the coverage ratio are removed and clear criteria are established for choosing between the IRB and standardized approaches. Governance, data and validation requirements are reinforced, and the definition of default is harmonized. The guidance details the estimation of risk parameters: probability of default (PD), loss given default (LGD) and credit conversion factor (CCF), requiring data quality, robustness testing and margins of conservatism (MoC).
• Market risk. Differentiated expectations are set for the CRR III framework, with clear guidelines for the transition. The guidance strengthens requirements for validation, documentation and supervision of value-at-risk (VaR), stressed VaR, expected loss, incremental risk charge (IRC) and risk management methodologies not included in the model engines.
• Counterparty risk. Stricter requirements are introduced regarding risk horizon (Margin Period of Risk, MPoR), initial margins, collateral and simulation granularity. The guidance defines expectations for parameter calibration, treatment of positive expected effective exposure (EEPE), justification of the alpha parameter and consideration of risks not captured in the EEPE. It also reinforces the use test and independent validation.

Download the technical note on the Revised Guide to Internal Models (EGIM)available in English and Spanish.