SUBJECTS

SUBJECTS
In-depth theoretical and practical know-how
SUBJECTS
In-depth theoretical and practical know-how

 

In the risk management cycle (identify – assess – manage – monitor), risk assessment is a crucial task. Quantifying risks is a pre-requisite for managing and monitoring them.

 

Pool Approach

As regulatory expectations continuously increase, costs are set to rise and the trend towards convergence between banks seems to be unstoppable. The ECB guide to internal models (October 2019) is offering a way out of this dilemma, by taking the opportunity to use rating models developed from pooled data into consideration. Relatively unknown throughout Europe, this so called Pool Approach has existed in Germany for over 15 years and has been performing impeccably for several of the largest German IRB banks.

When it comes to modelling low default portfolios, the lack of historical data and default experience strikes banks as the most troubling issue. This fact is especially true concerning wholesale portfolios such as large corporates, banks, project finance or commercial real estate. By joining a Pool Approach, banks contribute their data to a continuously growing Data Pool, which reflects the basis for the corresponding IRB-models.

The RSU develops and runs several IRB-approved models based on this cooperative Data Pool. As a result of data pooling, the dataset available for modelling low default portfolios is significantly larger than the individual datasets available to the institutions sharing this approach. This broad database of historical and even more important default data, provides the modelling teams at RSU with a solid base for differentiated models. A limited amount of data leads to a more general modelling approach, often fitting several asset classes into one model, whereas having access to an extensive amount of data allows for greater sophistication and more accurate modelling. Additionally, when it comes to defending models towards supervisors and auditors, improved statistical options are immensely supportive. Another very important aspect, which is quite obvious, is the fact that having access to a certain amount of good quality data can lead to high stability of internal models.

Joining a Pool Approach comes with a specific division of labor. RSU assumes responsibility for central modelling, extensive parts of model validation and all central IT tasks. The institutions take care of their part of the model validation as well as all internal processes and contribute wherever may be necessary and specified.

Credit risk assessment

To assess the credit risk of individual obligors correctly, institutions need to gather, structure, weight, and appropriately calibrate relevant information, which requires large amounts of data. RSU has collected a pool of wholesale financing data that is unique in Europe. On this basis, our more than 35 methodology experts continuously review, validate, and develop our systems.

RSU’s rating systems are based on the concept of a holistic risk analysis: both quantitative and qualitative information is used for the assessment, as well as potential warning signals and external influences. Analysts need to obtain and evaluate this information; RSU provides the methodology for structuring and combining it.

Risk Analyzer provides automatic credit assessments based on the statistical analysis of capital market information.

Credit risk does not only refer to whether or not a loan is repaid, but also to the loss sustained in the event of a default. Our loss estimation systems enable our clients to perform valid statistical estimates of this risk component.

Anticipatory management requires early warning of any perceivable declines in creditworthiness. Risk Guard analyses listed companies and identifies default risks sufficiently (up to one year) in advance.

 

Basel III/IV

Since 1 January 2014, the Basel III rules have applied to credit institutions and financial service providers operating in the EU. Basel III builds on the Basel II framework on capital standards. It requires institutions to maintain certain minimum levels of (regulatory) capital to cover credit risks, more specifically the unexpected loss from credit exposures. This amount depends on the one-year probability of default (PD), the exposure at default (EAD), and the loss given default (LGD) of each exposure.

To determine regulatory capital requirements, a bank can choose between the Standardised Approach and the Internal Ratings-based Approach (IRBA). The former provides for fixed risk weights and the use of external ratings, the latter requires institutions to rate obligors themselves.

There are a Basic and an Advanced IRB Approach. Under the Basic IRBA, probabilities of default are estimated by means of officially approved internal rating systems whereas LGD values and the method for determining the EAD are fixed. By contrast, the Advanced IRBA also provides for individual estimates of the LGD and of specific components of the EAD and is thus the most sophisticated method of determining capital requirements based on actual risk.

Pursuant to BTO 1.4 of the German Minimum Requirements for Risk Management (MaRisk), even institutions that use external ratings to calculate capital requirements (Standardised Approach) must adopt reliable systems for the initial and follow-up assessment of credit risks. This provision is part of the implementation of the second pillar of the Basel framework.

RSU’s internal rating systems are successfully used by many clients under the IRBA. They also meet the requirements of the MaRisk.

The same applies to RSU’s loss estimation vsystems, which have been used under the Advanced IRBA since 2008.

 

Solvency II

Since 1 January 2016, insurance companies have needed to determine their Solvency Capital Requirements (SCR) according to the Solvency II rules.
The risks of financial investments must be assessed und appropriate weights determined. This can be done using a standardised model, in which case most of the weighting parameters are fixed.
Using an internal model (component) allows insurers considerably more discretion in determining capital requirements, provided they fulfil extensive regulatory conditions.
Irrespective of the type of model chosen (standard or internal), insurers must assess their risks continuously and check the impact of theses risks on their financial situation (Own Risk and Solvency Assessment, ORSA, required under pillar 2 of the Solvency II framework).
According to the Prudent Person Principle, insurance companies need to understand the risks associated with their investments and be able to assess and manage those risks.
The above concepts are specified in the German Versicherungsaufsichtsgesetz (VAG).

RSU’s rating systems cover a broad range of asset classes. For each rating, the obligor’s risk situation is comprehensively examined so that investors can take informed decisions.
In addition, the rating systems are checked regularly and in detail by national and European supervisory authorities.

Using Risk Analyzer for the automatic monitoring of listed companies further enhances the risk assessment process.

 

IFRS 9

Under the rules of IFRS 9 for valuing financial instruments, due to take effect on 1 January 2018, the Incurred Loss approach will be replaced with a (Lifetime) Expected Credit Loss model. As a result, key metrics used in risk management (PD, LGD, EAD) will also become relevant for accounting purposes.

The new approach essentially provides for the expected loss of a financial instrument to be calculated over the instrument’s entire term. The Migration Matrices/PD profiles (MMX/PDP) complementing RSU’s rating systems enable users to determine the relevant long-term developments of their financial instruments.

 

Stress Testing

Both national (BaFin) and European supervisory authorities (EZB, EIOPA) require institutions and insurers to perform regular stress tests to ensure that capital requirements are met.
For this purpose it is necessary to quantify the effect of crises on future credit losses and credit quality.

RSU’s Stress Test Analyzer uses macroeconomic (stress) scenarios selected by the user to compute shift factors for the adjustment of default rates and PDs at country, industry, and asset class level.