Skip to Content

Implementing IRB slotting models - How to overcome the biggest challenges

Katie Hill, Head of Commercial Lending

The author

Katie Hill

Head of Commercial Lending

Adopting an Internal Ratings Based (IRB) approach is high on the agenda for many firms, but those with specialist portfolios face particular challenges in meeting the regulatory requirements. In this blog, Head of Commercial Lending, Katie, provides guidance to help you overcome them.

The European Banking Authority’s (EBA) recent publication on Regulatory Technical Standards (RTS) provides a much clearer structured approach on slotting models for relevant portfolios. 

So, what’s a slotting model?

Essentially, it’s a way to group accounts according to their level of credit risk where there is insufficient data to build traditional PD, LGD and EAD models. Accounts are allocated to one of five overall categories (Strong, Good, Satisfactory, Weak and Default) by assessing several points specified in the publication (called factors and subfactors).

These cover information such as the customer’s ability to pay, the valuation of the asset and its sensitivity to the economic environment, the ability of the organisation to foreclose, and the environment which the customer operates in.

It is recognised that there are differences by organisation and portfolio in the importance of factors and sub-factors so weightings are applied which must fall within a specified range.

How does this impact IRB?

Prior to this point, the PRA had stated that slotting models were the recommended approach for IRB but they gave no in depth guidance for how these models should be constructed.  The issuance of the framework from the EBA provides additional clarity although it does not mean that the process is without difficulties. From our experience of supporting our clients with slotting developments, we have identified three common challenges in these developments.

 

The three biggest challenges:

1. Data

The challenge

For commercial lenders, moving to IRB requires uniformity of data and processes to make consistent decisions about the credit risk of an account. Accessing the required data can be hampered by issues such as multiple legacy systems, paper-based decisioning or free-format data entry.

Using tools like Echelon enables you to process credit risk applications faster and with greater consistency, whilst gathering and storing data to enable a cycle of continuous improvement. Echelon enables you to apply consistent credit scoring approaches to all applicants, now enhanced with direct links to information on companies and an ability to monitor and explore high risk companies.

Top tip

That's why we always recommend that firms start by managing their data at the outset of the project  The development of slotting models often goes hand in hand with data management programmes and the implementation of systems to capture this information in a consistent manner.  We recommend, therefore, that you involve data experts from the earliest stages, during the initial planning, to ensure your data management process is fit for purpose from the start, thus ironing out any issues that would otherwise occur later.

2. Translating Framework into Models

The challenge

The guidance from the EBA provides a framework of Factors and Subfactors for inclusion in a development.  However, the missing part is translating the framework into something tangible. In some parts the guidance is prescriptive and in others it can be quite vague.  For example, a strong account would comply with “Property is located in highly desirable location that is convenient to services that tenants desire” whilst a good account would adhere to “Property is located in desirable location that is convenient to services that tenants desire”.

Sounds simple enough but the devil is in the detail! What’s the best way of translating what you know about a property location into its desirability?  How do you differentiate between highly desirable and just desirable?  Would this be reflected by the financial return on letting the property, is it the footfall in the specific location or is it the transport links meaning that customers can easily access the location?

Top tip

By their nature, slotting models are subjective although it is necessary to be as consistent in the assessment of a factor as possible (especially since many people might be making these assessments!).  Translating what is agreed to be a strong location (for example) into a clear definition of what constitutes this is necessary to negate variation. 

3. Post Development Continuous Evaluation

The challenge

Slotting models are designed to be updated and therefore the credit risk of accounts need to be refreshed with any new information that is available.  Our experience shows that much of the information in a slotting model for these types of portfolios is static.

But is it…?

Taking location from the previous example – a shop on the High Street in Leeds might be considered a highly desirable location at origination.  But let’s imagine some of the big stores are closing down – what does that mean for accounts in your portfolio now? How and when is the call made to downgrade these properties from excellent to good?

Top tip

A regular cycle of account level review is necessary to ensure that the slotting of an account remains as accurate as possible.  The development process is as much about how the model will be used, reviewed and updated as it is about the actual composition of the model.  The post development process needs to be considered early on and must be subject to governance and continual review to ensure compliance with the Capital Requirements Regulation (CRR).

 

Depth of data is no longer an issue

This was previously a challenge that prevented many lenders from transitioning to IRB. But, not any more! Depth of data is no longer a barrier because long histories and / or significant volumes of defaulted accounts are not required to develop slotting models.  What is required is a thorough understanding and articulation of your portfolio and the account characteristics and behaviours to indicate that there is a deterioration in Credit Risk. Whilst the guidance has been provided to developing slotting models, the challenges described show that it is by no means a simple process. 

How we can help

We have worked collaboratively with our clients to translate the EBA framework into weighted factors and subfactors which are relevant to an organisation’s bespoke portfolio. To find out more about this experience and our general commercial lending consultancy services, click here.

Or, to find out about our commercial lending product, Echelon, click here. Echelon, Jaywing’s commercial underwriting platform, is a browser-based tool that enables you to use slotting models in application processing and ongoing account review. It now includes links to Companies House data and features a unique ability to monitor, visualise and explore the high risk links between your customers and connected organisations.