On August 5, 2024, the Reserve Bank of India (RBI) released a draft circular outlining regulatory principles for managing model risks in credit. This draft aims to address potential risks and ensure robustness in credit management models used by regulated entities (REs) for tasks like credit appraisal, borrower scoring, and risk management. The circular emphasizes the need for comprehensive governance, oversight, and validation processes for these models. It mandates that REs must adopt a Board-approved policy covering model development, deployment, and ongoing validation. The draft also stipulates that new models must comply with these principles within three months of issuance, and existing models should be validated within six months. Comments on the draft are invited from stakeholders by September 4, 2024, and can be sent to the RBI’s Chief General Manager via mail or email. The new guidelines aim to mitigate model risks, enhance transparency, and ensure regulatory compliance. The circular repeals the 2002 guidance note on credit risk models, reflecting updated regulatory requirements.
Reserve Bank of India
Press Releases
Date : Aug 05, 2024
RBI releases draft circular on ‘Regulatory Principles for Management of Model Risks in Credit’
Regulated Entities (REs) generally use various models as part of their credit management, including for credit appraisal, borrower scoring, pricing, risk management, etc. With a view to addressing potential risks from use of such models and ensuring robustness in the process of model deployment by REs for credit decisions, the Reserve Bank of India today released the draft circular on ‘Regulatory Principles for Management of Model Risks in Credit’. The Comments on the draft circular are invited from public/stakeholders by September 4, 2024.
Comments/ feedback may be forwarded to:
The Chief General Manager
Credit Risk Group
Department of Regulation, Central Office
Reserve Bank of India, 12/13th Floor,
Shahid Bhagat Singh Marg,
Fort Mumbai – 400 001
or
by email
(Puneet Pancholy)
Chief General Manager
Press Release: 2024-2025/835
DRAFT CIRCULAR FOR COMMENTS
RBI/2024-25/
DOR.STR.REC. /21.04.048/2024-25
August 05, 2024
All Commercial Banks (including Regional Rural Banks and Local Area Banks)
All Primary (Urban) Co-operative Banks
All State Cooperative Banks & Central Cooperative Banks
All Non-Banking Financial Companies (including Housing Finance Companies)
All All-India Financial Institutions
Madam/Dear Sir,
Regulatory Principles for Management of Model Risks in Credit
Regulated Entities (REs) use various models as part of credit management life cycle for borrower selection, credit scoring/ rating, pricing, risk management, credit loss provisions etc. Inherently, model outputs are exposed to uncertainties as they are based on assumptions which may not manifest in the envisaged ways and may take different forms in a real-world scenario. This potentially exposes the REs to model risk, which has implications on prudential aspects of credit risk management, compliance and reputational risk. While the application of technology in models has facilitated faster decision-making under complex scenarios, it also adds complexity to the model risk management framework implying the need for a comprehensive understanding, a robust validation mechanism as well as appropriate governance and oversight.
2. With a view to ensuring prudence and robustness, in the use of such models, it has been decided to lay down certain broad regulatory principles as provided in the Annex which shall be adopted by the REs.
3. The circular shall come into force within three months from the date of issuance. New credit models to be adopted by REs shall follow these guidelines with effect from the above date and the existing models shall be validated in terms of these guidelines within six months from the date of issuance of this circular.
4. The above instructions are issued under Sections 21, 35A and 56 of the Banking Regulation Act, 1949, Sections 45JA, 45L and 45M of the Reserve Bank of India Act, 1934, and Sections 30A, 32 and 33 of the National Housing Bank Act, 1987.
Yours faithfully,
(Vaibhav Chaturvedi)
Chief General Manager
Regulatory Principles for Management of Models Risks in Credit
A. Definition:
1. ‘Credit risk model’ refers to any quantitative method that applies statistical, economic, financial, or mathematical principles and assumptions to process data into an output to be used for credit decisions.
Credit decisions, for the above purpose, shall cover all decisions involved in management of a credit exposure and may inter alia include credit scoring and borrower selection; pricing of loans; risk analysis of various loan categories; estimating loan loss provisions and economic capital.
B. Governance and oversight
2. Regulated Entities (REs) shall put in place a detailed Board approved policy (hereinafter referred to as ‘policy’) with regard to model risk management framework for all models deployed, covering the entire model life cycle. The policy shall cover, inter alia, details of governance and oversight aspects commensurate with model materiality; processes around model development or selection; documentation for models deployed; independent vetting / ongoing validation or review processes; change control; and the monitoring and reporting framework including role of internal audit function. The policy shall also cover the approach with regard to adoption and usage of third-party models. The REs shall maintain a Model Inventory of approved models, insourced or outsourced, with critical information on it.
3. The deployment of individual credit models adopted under the policy, and any subsequent changes in their inputs or assumptions, shall be with the approval of the Risk Management Committee of the Board (RMCB) or any other Sub-Committee, as designated by the Board.
C. Model development and deployment
4. The models used by the REs may either be developed internally or sourced from external third-party suppliers, including under collaborative lending arrangements, or can be a mix of both as per the provisions of the policy, subject to the following broad principles:
a. The objectives of the model, problem statements and solution sought from the model should be clearly defined and shall be essential pillars of model development.
b. The inputs and assumptions considered for the model development shall ensure adequate robustness, with a view to effectively address the intended objectives of the model on a consistent basis.
c. There shall be a detailed documentation for each of the models which should include the details of the sensitivity of the outputs to the assumptions and inputs, to facilitate clear understanding by RE’s users, its top management and supervisors.
d. The model should have the necessary scalability and flexibility to meet the needs of dynamic business conditions.
e. The model shall have the necessary interface with core banking/financial system, liquidity management, asset liability management (ALM) or any other risk management system of the RE.
f. Outcomes of the model shall be consistent, unbiased, explainable and verifiable. The same shall form part of the model validation framework.
g. In case subjective factors are used to override model outcomes, then the same shall be as per the provisions of the policy and such deviations shall be suitably documented in an auditable format.
5. In cases where outsourced/ third party models are used, the models shall mutatis mutandis be subject to above broad principles. The contractual agreements with the third parties shall provide for access by the RE to minimum technical documentation on the models being used or acquired that should give reasonable understanding on design, configuration and operation of the model. REs shall be ultimately responsible and accountable for the integrity and outcomes of outsourced models.
D. Model validation framework
6. REs shall put in place a model vetting / validation process, independent of model development / selection, for assessing the robustness of models developed inhouse or otherwise. Each model shall be validated before deployment as well as after any subsequent amendments owing to material events or as part of periodic reviews, which shall be at least on a yearly basis. The REs may also consider engaging external experts for validation of the models deployed by them, as provided under their policies.
7. The validation exercise shall inter alia include review of assumptions underlying the model for their validity / substantiation; verification of the accuracy of data used in the model and reliability of data sources; confirming compliance with extant regulatory / statutory instructions; evaluation of model documentation for completeness and accuracy, and assessment of the efficacy of the model outcomes through back-testing. Besides, the validation exercise shall comprehensively review all the limitations and weaknesses in the models, including instances of bias or discrimination, if any, to ascertain the need for suitable corrective measures.
8. Validation outcomes shall, inter alia, be in terms of suitable and easily understood ex ante parameters. The results should be compared against benchmarks prescribed in the policy framework and should be placed before the Risk Management Committee of the Board (RMCB) or the designated Sub-Committee of the Board as mentioned at clause B above.
9. The models deployed by the REs shall be subjected to supervisory review. The Reserve Bank may also engage external experts to validate the models deployed by REs, including the external models deployed for their credit management, based on supervisory risk perception. Accordingly, wherever REs have engaged external models in terms of clause C above, such arrangements shall include appropriate contractual provisions enabling supervisory evaluation of such models either directly by the Reserve Bank officials or by other external experts as engaged by the Reserve Bank.
List of Instructions repealed
| SI No | Circular Number | Date of Issue | Subject |
| 1 | Guidance Note on Credit Risk Management (Chapter 3 – Credit Risk Models alone repealed) | October 12, 2002 | Credit Risk Management |
