Credit Risk Management

Print this page Create a PDFShare this page
Send to a friend
Credit Risk and Counterparty Credit Risk Modeling, Validation, Software Solutions and Consulting

Credit Risk

Credit risk encompasses default risk (for instance, a debtor has not met his or her legal obligations according to the debt contract, see National Bank Regulations tab), migration risk (stemming from adverse movements in internal or external ratings) and country risk(he/she cannot per perform or pay because of events or measures taken by political and monetary authorities).

With respect to Basel Regulation, many banks chose to develop their own estimates of credit risk parameters: Probability of Default (PD), Loss Given Default (LGD), Exposure at Default (EAD). Business & Decision Risk Technology Consulting has gained solid experience by developing models for the Internal Ratings Based Approach (IRBA) at different customers. With proven years of experience in the effective implementation of quantitative models, Business & Decision can assist you in the requirements, development, recalibration and validation of credit risk models.

For the implementations we used tools such as:

  • SAS Credit Risk for Banking
  • SAS Enterprise Miner & SAS Credit Scoring
  • Matlab

Probability of Default Curves for CounterpartiesInformation text on the opening of the image in lightbox (magnifying glass)

Credit and Counterparty Risk Validation

Models developed for the risk computation have to be revalidated regularly.

On one side, the second pillar of Basel regulation implies that supervisors check that risk models are always performing consistently. On the other side, recent crisis has drawn the attention of internal stakeholders of the bank (business, CRO) to a higher interest on the models.

The validation process consists in a review of the development process and all related aspects of model implementation. It can be split in two parts:

  • Quality control worries about the ongoing monitoring of the model use, the quality of the input variables, the judgmental decisions, and the resulting model output.
  • Quantitatively, the backtesting compares statistically the predicted risk parameters with the actual outcomes.

 

In the credit risk context, the validation process concerns the three parameters: probability of default (PD), the exposure at default (EAD) and the loss given default (LGD). For each of those three, a complete backtesting is done at three levels:

  • Discriminatory power: is the model able to differentiate between defaults and non-defaults, or between high losses and low losses?
  • Predictive power: this is check by comparison between the estimates and the realized values
  • Stability: did the portfolio changed between the moment when the model was developed and now?

In this three x three matrix (parameters x levels), each component has one or more standardized test to process. Business & Decision Risk Technology Consulting can implement those tests and provide all needed reporting.

In the counterparty credit risk context, one must take into consideration the uncertainty of exposure and the bilateral nature of risk. Hence, exposure at default is replaced by expected positive exposure (EPE) and effective expected positive exposure (EEPE). The test consists in comparing the observed P&L with EEPE (violations should be moderate, and the “pass” rate should not exceed a predetermined level, for instance 70%. For illustrations, we provided the following example:

 

Counterparty Risk Profit and Loss Simulations

Counterparty Risk Positive Exposure Simulations

 

 

 

 

 

 

 

 

 

 

 

 

 

Counterparty Risk Expected Exposures

Model Risk

The Belgian regulator, the National Bank of Belgium (NBB) insists that appropriate conservative measures be taken to compensate for the shortcomings of value and risk models. For instance, the NBB requires that there be an assessment of model risk, based on an inventory of:

  • The risks covered by the model, together with an assessment of the quality of the results calculated by the model (maturity of the model, adequacy of assumptions made, weaknesses and limitations of the model…) and the improvements planned over time.
  • The risks not yet covered by the model, an assessment of the materiality of these risks, and the way they are handled.
  • The entities that are covered by the general modeling methodology, as well as the entities covered by a more simplified methodology, or the ones that are not covered at all.

Business & Decision Risk Technology Consulting can provide you the necessary functional and technical knowledge to evaluate the model risk.

Why B&D?

With proven years of experience in the effective implementation of quantitative models, Business & Decision Risk Technology Consulting an assist you to manage models from strategic, usage and operational angles. More precisely, Risk Technology consultants can help you to:

  • Ensure that methodological and modeling items are consistent across your organization
  • Assess that models developed internally comply with corporate business objectives
  • Check if model monitoring policies are coherent and in conformity with the model
  • Update model documentation to make it transparent for business continuity purposes
  • Ensure that models remain minimally in line with market standards
  • Enforce traceability of all the model lifecycle steps with a model inventory

 Business & Decision Risk Technology Consulting can assist with the technical and functional expertise in the key points of model development:

  • Initiation
  • Improvement
    • Request
    • Design
    • Prototyping
    • Model Analysis & Documentation
  • Model Management Cycle
    • Review
    • Validation

Case Studies

Credit Risk CalculationsCredit & Counterparty Risk Calculation

Case Study - Documentation of the Credit Risk Model for Internal Model Approval Process

Partners


Want to know more?

Contact Business & Decision Risk Technology Consulting

2016 Update ECB's AnaCredit Regulation

To get to know the challenges of the ECB's AnaCredit Regulation and how Business & Decision proposes to address them practically:

Click on the image to download the white paper

Be Prepared for ECB’s Analytical Credit Dataset 2016 Business & Decision Benelux