In its effort to reduce the variability in modelled Risk Weighted Assets (RWA), the BCBS has included a Standardised Approach Floor to IRB RWA results in the Basel IV framework. The SA floor will be implemented as of 1 January 2022, with a 5-year phase-in period to smoothen out the impact. At the moment, initial impact estimates indicate a large impact on capital requirements, while the impact on financial institutions’ operations is still being debated. In any case, the addition of the SA Floor is making huge waves. In this article, Empyrean Solutions examines additional effects of the SA Floor and what challenges lie ahead.
One of the key additions in the Basel IV framework is the SA floor as a back stop to the reported IRB RWA’s. The main objective of the floor is to promote the comparability of RWA’s across IRB banking institutions by limiting excessive variability of modelled RWAs. It is also believed that the backstop might trigger a more conservative approach towards modelling going forward by reducing the capital relief potential of the IRB approach.
The impact of the output floor on capital requirements is material as EBA’s October Basel III monitoring report showed a 19.3% increase in Pillar 1 capital requirement after the full Basel III reforms are implemented. The capital impact also seems to be stable overtime hovering around 20% across the 3 data sets used in the exercise (Dec 2017, June 2018, Dec 2019). The monitoring exercise also shows that 5.4% if the overall capital impact can be attributed to the Output Floor, making it the single largest driver.
This massive impact on IRB bank’s RWA numbers explains the multiple EU-US negotiation rounds on the topic of the output floor before it was penciled in as part of the final package. Geo-political negotiations between US and EU supervisory bodies on the level of the output floor delayed the process. This reflects the differences in the impact of the output floor on the competitive positions of the IRB banks on both sides of the Atlantic. EU’s initial position was that 70% was the maximum acceptable level, while the US insisted on a higher level (the Collins Amendment to the Dodd Frank Act introduced a floor to the US banks in any case). The ultimate outcome at 72.5% was generally assumed to be a challenge for EU IRB banks, which seems supported by the impact from EBA’s study.
Apart from the question on the competitive level playing field aspect, there are other concerns banks might have in relation to the SA output floor.
Firstly, the output floor might trigger banks’ senior risk management to reflect on the question whether the cost of maintaining some of their (expensive) IRB models is still worthwhile, given that their capital relieve impact became at least partially irrelevant. This outcome would mean an unsolicited side-effect of the output floor that goes directly against the overall objective of promoting modelling across the board.
Secondly, the mechanics of the output floor need to be considered carefully given the aggregation level of the floor. In a BAU context, this aggregation introduces an unintended complication since the marginal contribution to regulatory capital requirements of an individual deal or (sub-) portfolio is no longer known after the (aggregate) floor kicks in. This might complicate BAU business decision making based on cost of (regulatory) capital parameters.
Thirdly, a similar challenge lies in the interpretation of bank level stress testing results in a context where the output floor applies. Stressed risk factors have different impacts on capital requirements in the SA and the IRB approach respectively. Therefore, any outcomes of stressed scenarios are uncertain given that both the IRB capital requirement and the SA floor will react to the stressed conditions differently. Similar to BAU, impact analysis of stress scenario’s to sub-portfolio levels will equally become complicated in cases where the output floor applies. A similar effect was also described by ECA in its conclusion of the 2018 stress test “[…] the bottom-up approach was constrained by imposing a number of caps and floors […]. The EBA did some ad-hoc assessments […] to estimate the impact of these caps and floors on the banks’ results. However, given its limited resources, the EBA has not been able to produce a comprehensive overview of the impact of these caps and floors on the results.” Even though the ECA refers to the overall lack of transparency on how caps and floors generally impact bottom-up results, it is crystal clear that the introduction of the SA output floor has pushed the issue center stage.
Finally, the EBA already pointed out potential challenges and overlaps between the output floor and Pillar 2 stating that it is “[…] important that Pillar 2 and systemic risk buffer decisions are reviewed in a framework that includes the output floor […]”. In its final conclusion, the EBA recommended the use of the floored RWA for both Pillar 1 and Pillar 2 requirements and insists that national competent authorities revise their Pillar 2 requirements taking the impact of the Output Floor into consideration. However, EBA refrains from providing guidance on the ‘how’-question.
In conclusion, while overall the output floor meets its immediate objective of serving as a backstop against RWA variability, more work lies ahead to work around some of the unintended challenges that it brings with it. Unintended consequences and operational complexities need to be sorted out. Banks should therefore anticipate further regulation to modelling that complement the output floor, also given that loan loss provisioning under IFRS9 seem to suffer from an even greater level of output variability. Even though the most recent Basel framework is based on what is called ‘the finalisation package’, further work is required on the modelling part of the framework. Banks defining their strategies for compliance with the international regulations like Basel and IFRS should keep the ever-evolving nature of this area of the regulation in mind and make sure their systems and processes allow for a sufficient degree of flexibility and adaptability. In addition, banks should be able to easily assess impacts to model changes to capital and provisioning alike.
Join Piet Mandeville, Product Director - Risk, Empyrean at RiskMinds International! You can catch him on the stage at the Global Risk Regulation Summit.
 EBA, Basel III Monitoring Exercise, Oct 2019
 From the Governors and Heads of Supervision (GHoS) press release, which was published together with the December 2017 revised framework, it was clear that the floor is a single, aggregated floor: ‘The reforms endorsed by the GHoS include the following elements: […] an aggregate output floor, which will ensure that banks’ risk-weighted assets (RWAs) generated by internal models are no lower than 72.5% of RWAs as calculated by the Basel III framework’s standardized approaches […]’.
 European Court of Auditors’ Special Report “EU-wide stress tests for banks: unparalleled amount of information on banks provided but greater coordination and focus on risks needed”, 10 July 2019.