Challenged by heightened geopolitical risk and spurred by the pending finalization of Basel III, EU banks should reconsider their capital optimization. While there are many factors to consider, the choice between the standardized approach or internal ratings-based (IRB) models for credit risk will heavily influence capital strategy.
Geopolitical risks, of course, are reshaping the world – whether we’re talking about ongoing wars between Russia/Ukraine and Israel/Hamas, the recent military conflict between Iran and Israel and the United States, AI-driven government espionage, or terrorist-led attacks on commercial shipping vessels in the Red Sea. The fallout from these events impacts everything from market risk and credit risk to operational resilience. Indeed, they have as resulted in jittery financial markets, as reflected in extremely high values of the VIX/Uncertainty index in April this year.
Marco Folpmers
Complicating matters further is the pending finalization of Basel III (aka the Basel Endgame), the international regulatory accord that includes reform measures (e.g., the CRR3 directive) that are set to increase bank capital requirements. In addition to modified input floors for capital calculations and a new output floor, banks will also have to contend with closer scrutiny from supervisors during on-site inspections.
Under these circumstances, it makes sense for banks to rethink their capital strategy. Is an IRB model, for example, still sufficiently valuable? Just as importantly, what are the current the total costs of ownership of IRB, if one considers the costs generated by all modeling and validation activity, the data teams, senior management involvement, project management, training, agile scrum masters, and model monitoring and model risk management?
The alternative to IRB is the standardized approach (the fallback approach given by the Basel Committee on Banking Supervision), which comes with an increased capital requirement. If a bank finds that its IRB models no longer provide sufficient capital relief, one could argue that it is better for that institution to just implement the standardized formula and to reduce its quant workforce – possibly re-allocating them to other purposes, like commercial pricing models.
Let’s now unpack the complicated IRB vs. standardized approach debate and examine its far-reaching consequences for banks.
A Complicated Decision
Whether a bank keeps or transitions to an IRB model (coming from the standardized approach) or shifts/simplifies from IRB to the standardized approach is a complex decision – and should not be made hastily.
When deciding between IRB and the standardized approach, every bank should start with two important and overarching questions: what is our capital strategy and what do important external stakeholders expect from us?
The capital planning discussion starts at the board level. Does a bank, for example, desire a modest risk profile, limiting itself to safe and well-known credit portfolios? Or is it willing to take on more risk and invest in more risky clients (e.g., innovative firms and start-ups), aiming for a higher risk/reward strategy?
At this stage, the next set of questions can be tackled: e.g., how is the bank’s risk appetite reflected in its capital ratios? What target values does it have for the CET1, total capital and leverage ratios? And how do these target values relate to the supervisory minimum values for these ratios – i.e., is there extra capital, or “headroom,” at hand?
If there is headroom, one could (prematurely) argue that the bank can switch some of its portfolios from IRB to standardized, absorb the additional capital requirement in the current headroom (which will either be reduced or even totally disappear), and benefit from the lower costs involved in the application of the standardized approach.
Alternatively, one could argue (again, prematurely) that if a bank were to favor the IRB approach, it could expand its balance sheet, increase its profits and use up the headroom for extra business, exposures and income generation.
However, after a bank has identified its extra capital, it may be better off just hitting the pause button. The headroom can only be spent once, after all, so perhaps it should be allocated to other purposes – e.g., set aside for inorganic growth or distributed to current investors.
Moreover, there may be other important parties (outside of the boardroom and senior management) who have ideas about this headroom. To help us determine whether IRB or the standardized approach would work better for a bank, let’s now examine the wants and needs of these external stakeholders.
External Stakeholders
The market at large is the first “external stakeholder.” The market’s trust in a bank’s health is reflected in its stock price – and especially its price-to-book ratio. The latter ratio (which is typically low for banks) is multi-layered. For the purposes of this article, though, it’s sufficient to say that if the price-to-book ratio is low and/or declining, a bank should be very hesitant to use up potential headroom for a switch to the standardized approach.
Rating agencies, like Fitch, S&P and Moody’s, are the second group of external stakeholders. The external rating that they attach to a bank’s debt is extremely important for the bank’s capital management – as is the agencies’ motivation for this external rating.
If a bank targets a specific rating (say, AA) and the rating agency reports that the bank has sufficient capital and earnings potential to protect that rating, that is fine. However, it is important to look under the hood and to consider under what circumstances these agencies would consider a downgrade. If a bank’s CET1 ratio, for example, is slightly below the value expected by the rating agency, it actually has no extra capital, and any plans to use up this headroom would put it at risk of a downgrade.
Rating agencies, moreover, look carefully at pillar 2 economic capital, so a bank must have sufficient capital to cover that requirement. Simply managing risk-weighted assets (RWA) is not enough.
Let’s now consider the pros and cons of both the standardized and IRB approaches for this group of stakeholders.
The Business Case for Shifting to a Standardized Approach
If there is still headroom after external stakeholder perspectives are assessed, IRB may no longer be attractive. Under this scenario, standardized approach may be preferred – to reduce operational costs.
What’s more, under the final version of Basel III, the new output floor and the higher input floor may cut into the relative attractiveness of IRB. However, IRB should remain a “no-brainer” for some portfolios – including sizeable, very low-risk portfolios like high-quality residential mortgages.
For other portfolios, such as SMEs or corporates, one may need to reexamine the costs and benefits of an IRB model. Let’s now consider the business case for either keeping or switching to the IRB approach.
The Business Case for Keeping or Switching to IRB
A switch to IRB could provide quantitative benefits in the form of reduced RWA – since it would accommodate a transfer from core capital funding to funding through (subordinated) debt. Assuming a required return on bank capital of 10%, and a coupon rate on the subordinated debt of 2%, the reduction in funding costs under the IRB model would amount to 8% of the capital.
Another way to assess the benefits of a shift from the standardized approach to IRB is by assuming the growth of the balance sheet that would be enabled by freeing up capital. This assumes that the bank is scalable and can easily expand its current business proportionally.
Of course, if a bank were to switch a portfolio from IRB to the standardized approach, the reverse would be true. Under that scenario, the capital requirement would grow and cheap funding costs (say, through subordinated debt at 2%) would be traded for expensive funding through investors – who would require (as in our example, above) at least a 10% return.
The finalization of Basel III is yet another factor that must be pondered when weighing the pros and cons of IRB models. The Basel III framework includes an output floor provision, which states that when a bank uses IRB to calculate its risk-based capital requirements, its RWA calculation cannot be lower than 72.5% of what it would be under the standardized approach.
So, if a portfolio is currently under the standardized approach but the bank is considering a switch to IRB, it should first check to see if there is room to reduce its RWA – to comply with the output floor requirement. If the bank’s RWA is only slightly above the floor of 72.5% of the standardized approach, it does not make sense to bring an additional portfolio under the IRB approach.
Annual cost is another factor to consider when deciding between the IRB and standardized approaches. For IRB, the number of employees (aka full-time equivalents) dedicated to model development must be considered. The cost can be calculated by multiplying all FTEs working directly or indirectly for the IRB model by the (weighted) average cost of labor.
The Importance of RWA Density
Comparing RWA amounts is the quantitative component of the decision between the standardized and IRB approaches.
If an IRB model does not exist yet for a certain portfolio, a bank can use RWA density statistics reported by the European Banking Authority (EBA) to gain a good idea about its potential impact. (These statistics are reported for the EU, as a whole and per bank.)
By selecting peer banks with a comparable risk profile and portfolio breakdown, one can get a good impression of the RWA densities – for both the standardized and IRB approaches. In the figure below, we show the overall RWA densities for three selected portfolios, across all banks within the EU’s EBA scope.
RWA Densities for Corporates, Retail Mortgages and (Corporate) SME
Source: EBA report, June 2024
We can conclude from the figure that RWA reduction via IRB treatment is substantial for the portfolios (corporates, retail mortgages and SME) under discussion. In short, this means it’s logical to retain IRB models for sizeable portfolios. Keep in mind, though, that the maintenance of an existing IRB model is easier than building one from scratch, especially when attempting to explain a methodology to a supervisor.
After comparing the RWA densities, one shouldn’t forget to apply some refinements to the business case. For example, a bank that migrates from the standardized to the IRB approach will benefit from lower funding costs. However, that benefit will be slightly reduced because of a second-order impact of the IRB switch: a thinner capital base may lead to higher debt-funding costs – even without a ratings downgrade.
The constraints of Basel III’s output floor and minimum leverage ratio must also be considered when weighing the pros and cons of switching from the standardized approach to an IRB model.
Parting Thoughts
Geopolitical risk and the Basel Endgame have recently grabbed the spotlight, motivating savvy banks to rethink capital management – particularly with respect to whether to employ the standardized or IRB approach to credit risk.
The heads of business units at a bank often want to focus on the “business case” for IRB and standardized models when deciding which approach would work better for capital optimization. However, the finetuning of the capital position is not just about RWA densities. Rather, it’s a much more strategic challenge that revolves around the bank’s risk profile and portfolio development, supervisor expectations, and investor and bond-holder preferences.
Dr. Marco Folpmers (FRM) is a partner for Financial Risk Management at Deloitte the Netherlands.
Topics: Modeling
