Risk Consulting
The Risk Consulting team advises financial players on credit modelling & data management.
We support banks and financial institutions in the following areas:
Credit modelling
- In-depth understanding of regulatory requirements and expertise in statistics to provide clients with methodological, and implementation supports for modelling, stress testing analysis, model validation
- Risk monitoring and maintenance reporting
Integration of ESG criteria into their credit model
- Helping Banks identify/ review their risk politics and risk governance structures
- Combining our knowledge with market best practices to propose the most suitable approach for defining KRIs for ESG risk appetites
- Defining and assisting our clients on the ESG integration roadmap
- Providing our expertise in ESG and credit risk quantification methodology and regulations as well as in ESG database
Data Management
- Designing data models for the different client objectives
- Providing ESG database and models to quantify ESG risks
- Improving data governance and quality
- Providing technology consulting services to integrate/ implement new IT infrastructures for risk management
Client benefits
A combination of our knowledge and market best practices
EthiFinance consultants propose the most suitable approach to define KRIs, providing the most appropriate assessments and recommendations regarding business allocation, portfolio management framework, risk and human resources
Credit risk measurement & analysis
With a in-depth understanding of regulatory requirements and expertise in statistics, our consultants provide clients with methodological and implementation supports for modelling, stress testing analysis, model validation as well as risk monitoring and maintenance reporting
ESG integration in credit risk
We support our clients on the ESG risk integration roadmap, providing our knowledge of the methodology and regulations in ESG and credit risk quantification as well as in the ESG database.
Mission on all Basel models (PD, LGD, EAD) and on all asset classes
Use Cases
Group Risk Management
Client
A leading European bank, offering its corporate, institutional and individual customers a wide range of value-added advisory services and financial solutions.
Mission
Modelling of the PD PRO/SCI Basel retail parameter
Objective
To propose a relevant, simplified, sustainable and fully compliant credit risk model.
Approach
- Construction of the modelling database (RDS)
- Risk differentiation: construction of the score model and classification of risk levels
- Risk quantification: calibration of the model
- Supply of the modelling package.
Implementation of the Backtest of Private Banking PD
Client
The private bank of a major European bank.
Context
As part of the monitoring of econometric models, the implementation of backtests of all credit risk models is a regulatory requirement. In particular, the PD model for private banking is a so-called expert model due to the very low statistical and quantitative involvement in its construction. Consequently, setting up a quantitative backtest was a challenge.
Objective
To meet a regulatory requirement. At the time of this backtest, we also had a second objective, which was to put in place new statistical monitoring indicators in line with regulatory expectations.
Approach
- Construction of the operating database. This database is a consolidation of several pieces of information from different universes to be merged via joins and controls
- Adaptation of the database for use in work tools
- Construction of indicators or equivalences based on the information available
- Implementation of regulatory statistical tests (PSI, HHI, AUC, GINI, binomial test, etc.)
- Integration of the validation team’s recommendations
Results
At the end of the backtest, the analyses confirmed that the PD values already implemented were conservative. As a result, no real action was required. However, a new model was being developed for this scope.
Contribution to the team
Carrying out this backtesting exercise allowed us to provide the modelling teams with a backtesting report in line with the regulator’s expectations. In addition, the teams have benefited from automated programmes that can be used on other subjects, thereby increasing productivity.
