2018 Stress Test Results: Banks look safe under no deal Brexit, but are the models right?
How to manage the impact of the PRA’s new stress testing principles
Transitioning to IRB: The challenges and benefits
14 November 2018
At around 10%, the number of UK lenders using IRB to calculate capital requirements in the UK market is small, but that figure is expected to rise significantly over the next few years.
Although the top-tier banks and building societies have either already implemented IRB, or are in the process of transitioning, many lenders are just starting out on their IRB journey.
So, what are the likely challenges – and benefits – they will experience? Ben Clark explores further in this blog.
What’s IRB all about anyway?
The internal ratings-based (IRB) approach is a bottom-up methodology used to calculate a firm’s credit risk capital requirement. It is driven by bespoke models to determine capital demand and is the alternative to the top-down, standardised approach (SA).
There are two approaches within IRB:
- Advanced IRB - where both the Probability of Default (PD) and the Loss Given Default (LGD) models are internal and bespoke. This is the most common approach used in the UK.
- Foundation approach - where the PD model is bespoke and the LGD model is prescribed.
Two key benefits
In a recent webinar with UK Finance, we discussed how IRB uses internal credit risk practices to estimate unexpected loss and is credited with bringing two key benefits to the banking industry:
- Improved risk sensitivity in capital requirement
- Better risk management practices within institutions
By incentivising better risk practices, IRB creates a safer financial system. But implementing IRB is a long process that brings with it many significant challenges.
What are the main challenges of IRB?
At Jaywing, we see four main challenges around implementing IRB. None of them are insurmountable, but businesses should be aware of them before they begin their IRB process:
For IRB models to work successfully, you need a lot of historic data. Whatever your current size or loans portfolio, you will need to develop a Through-The-Cycle (TTC) PD model and a Downturn LGD model.
The current UK requirements state that, at least for residential mortgages, the TTC PD model should reflect data as far back as the 1990s downturn. The current EBA guidelines suggest you need at least 20 years of data for the Downturn LGD model.
Most lenders simply don’t have this information available internally, so they need to use external data. For smaller banks and challenger banks, portfolios that have been built up over the last few years don’t have a sufficient volume of defaults and losses to model on.
However, there is some good news: new regulations from the PRA and EBA have clarified on how to use external data to build the required models.
A successful IRB development programme requires substantial knowledge, expertise and experience. A few years ago, the process of transitioning to IRB involved building a huge team, including modelling experts and second-line experts. It also required upskilling of management team to understand models, while data management and infrastructure expertise were both needed to implement models and use them successfully.
However, thanks to IFRS 9 many lenders have already substantially upgraded their modelling, infrastructure and data governance expertise in the past year or two, so the step up from where they are today to where they need to be for IRB standards is much smaller.
Similarly, most management teams within banks are now well aware of the PD, EAD and LGD frameworks. Linking these together and understanding that the framework is very similar is a big step towards IRB.
Costs and uncertain benefits
The costs of IRB are likely to be high. Often, these costs can be accurately estimated at the start of the project. However, the capital benefits are less certain because they do not become clear until the models have been developed. This situation presents inevitable difficulties in the decision-making process to implement IRB.
However, it is possible to split the project into phases and reassess the benefits after each stage. This allows for the capital required under IRB to be gradually refined, so decisions around when and how to continue to the next stage of the project can be based on realistic cost expectations. This prevents wastage if, at any point, the decision to transition to IRB is changed. Instead, work done up to that point can be put to use in improving IFRS 9 models, ICAAP processes and data governance, for example.
An IRB project takes at least three years to get approval and begin to achieve the expected capital benefit. And this is a minimum timeframe – regulatory challenges can further delay a project’s successful completion. Persuading a business to incur upfront costs for long-term benefits, rather than those immediately apparent, can be difficult.
The six benefits of transitioning to IRB
So, with those difficulties to overcome, what are the main attractions of transitioning to IRB?
1. Capital saving
This is the most obvious benefit of making the move to IRB. Capital savings are generally large although the benefit is likely to be impacted by the capital reforms outlined in BCBS 424.
2. Improved governance
By achieving IRB, you will implement a better governance structure around modelling and decision-making. That, in turn, will allow you to make better decisions across your organisation, have better models in place, and have fewer risks.
3. Internal expertise
You will develop a better understanding of your risk profile as you go through IRB. You will have to analyse your risk profile in ways you haven’t done before, developing your models and understanding where you need to apply conservatism, as well as understanding where your risk profile will be similar or different to the rest of the industry. Understanding all of that will help you to make better lending decisions in the future.
4. Model and system synergies
As well as similarities with IFRS 9, there are also synergies with stress testing and making lending and management decisions. By bringing these models together, you can achieve significant resource and cost savings from the modelling team and governance processes.
5. Sign of sophistication
Being an IRB bank increases confidence and trust in your ability to manage your risk and your management practices. It shows that you’re a well governed, well run, well organised bank – and that stamp of approval can lead to both reduced funding costs and positive impacts on share prices.
6. Improved data management
As you complete the IRB process, you will have to improve your risk data management and data governance processes. The good data practices that come with transitioning to IRB will help your business benefit such as improving the predictive power of models and the accuracy of reporting.
Seeing the bigger picture
IRB is not simply a capital saving project. It’s about changing the whole way you view models and data within your infrastructure. Yes, the big benefit is a significant capital reduction – but that’s not the way we at Jaywing would envisage starting an IRB project. You need to look at the whole gamut of benefits to understand its full potential, and our experts are able to guide you through each stage of the process with tailored guidance and assistance.
For additional guidance on navigating the challenges and how to benefits from, download our whitepaper.