The final leap: implementing the Basel III reforms in Europe

Speech by Elizabeth McCaul, Member of the Supervisory Board of the ECB, at the Working Group Financial Services hosted by Kangaroo Group on “Finalization of the Basel III Framework - a view from the BCBS”

Frankfurt am Main, 8 September 2021


Thank you for inviting me to speak today. It is a pleasure to join this discussion as the Kangaroo Group’s main goals to deepen the European Union are very appealing for a representative of an EU institution. The objective of a full implementation of the EU Internal Market, where the Kangaroo group has been instrumental since the early 1980’s, is particularly dear to our hearts. We should also nod to the significant role played in the launch of the euro by the former Secretary-general of the Kangaroo group, Karl von Wogau, as then chair of the Parliament’s ECON committee in the late 90’s - and, coming from the European Central Bank, the aim of a stable single currency is indeed equally dear.

Today’s topic is of major importance for the European banking sector, the European Commission and the ECB as a banking supervisor. I would like to discuss whether the coronavirus (COVID-19) pandemic should have an impact on the timeframes for the Basel III reforms, and present our views on the output floor, which is the most debated substantive component.

Not surprisingly, I will argue that full and timely implementation in Europe of the last set of Basel III regulatory reforms, without any further delays or changes in substance, is in the interest of all stakeholders. In fact, I wish we could take a big leap and jump like a kangaroo across the last mile of a process that began already some time ago after the great financial crisis of 2008.

Basel III reform and EU process

The agreement reached in 2017 by the Basel Committee on this last component of the reforms concerns the core of prudential banking regulation. It relates to promoting consistency and trust in the calculation of risk-weighted assets (RWA), and it strikes a sound balance between standardised and internal model-based approaches. The reform increases the robustness and risk sensitivity of the standardised approaches for credit risk, credit valuation adjustment (CVA) risk and operational risk, and it restricts the use of bank internal models by:

  1. removing loss given default modelling for low-default portfolios (namely banks and large corporates) and credit risk modelling for equity portfolios;
  2. introducing several model input floors in the area of credit risk;
  3. removing the use of internal models for operational risk and credit valuation adjustment risk; and
  4. adopting the much-discussed output floor, which requires that RWAs resulting from internal models not be less than 72.5% of the RWAs deriving from standardised approaches.

Following the onset of the COVID-19 pandemic, the Basel Committee decided to postpone the implementation date by one year from the beginning of 2022 to the beginning of 2023 to avoid contributing to business cycle disruption. This means that it will be at least 2028 before we achieve full phase-in. I support the Commission’s intention to initiate the implementation process in 2021 through a legislative proposal. It is important for the proposal to be launched as soon as possible in the autumn to enable the Council and the European Parliament to agree and pave the way for a 2023 European implementation. This timeline is certainly tight, but it should be followed so the new standards can apply fully in 2028 after the five-year phase-in agreed by the Basel Committee. Some in the banking industry may oppose this timeline as well as the substance of the reforms agreed by the Basel Committee, arguing that the impact may adversely affect banks’ capacity to support the recovery from the COVID-19 pandemic shock.

Recovery and reform

To be sure, uncertainty remains about the effects of the pandemic on bank balance sheets. While, economic projections have increased lately, and short-term risks have somewhat receded, uncertainty is still elevated. The need to manage loss absorption is an ongoing responsibility of banks – in this regard, banks increased capitalisation during the crisis overall and generally have the loss absorbency capacity to handle a further increase in provisions which can occur when public support is phased out. This helps to address uncertainty.

On the other hand, we see medium-term vulnerabilities that may yet materialise. The last leg of the Basel reforms plays an important role in ensuring our regulatory framework is fit for purpose to tackle any medium and longer-term vulnerabilities. The remaining reforms are proposed for longer-term systemic and structural reasons well-grounded in the lessons learned from the Great Financial Crisis.

It is worth looking at the arguments opposing the timeframes for Basel III implementation in Europe over both the longer implementation horizon and in the short term.

There is no rationale for opposition in the short term, as shown by three data points worth observing.

First, according to the European Banking Authority’s (EBA) most recently published impact study, while the impact of the Basel III reforms is heterogenous across the sample of banks, most are currently already able to meet the new requirements. So, it seems likely that they would be able to maintain lending to the economy even if some further COVID-19 losses materialised in the short term.

Second, the European banking sector is currently striking a positive tone in its forward guidance statements, as also evidenced by plans for dividend distributions. Indeed, the economic outlook for Europe is favourable: real GDP in the euro area is expected to grow beyond its pre-pandemic level during the first quarter of 2022. We have seen broad-based corporate activity pick up in the first half of 2021 – this trend is likely to continue and corporate profits are expected to reach pre-pandemic levels before the end of 2021.

Third and finally, the recent ECB analysis[1] of the macroeconomic impact of implementing the last leg of the Basel III reforms using alternative economic scenarios with varying effects from the COVID-19 pandemic shows that the short-run phase-in costs of the Basel III reforms in terms of GDP growth losses are outweighed by long-term resilience gains. This conclusion does not change under various COVID-19 impact simulations, and the one-year implementation postponement to 2023 helps neutralise negative effects on bank lending.

For the longer-term picture, we should keep in mind that Basel III was devised to pursue important long-term structural goals with two key objectives.

First and foremost, to make the banking sector more resilient. There is no room for complacency on the need for resilience; in fact, the pandemic provides a case in point. We were able to face the pandemic effects precisely because regulatory reforms over the past decade placed the banking sector in a strong enough position to manage the crisis, an external shock of a magnitude not seen before in Europe.

Second, to preserve an environment that supports credibility in the international banking markets. Multilateral agreements ensure a proper functioning of global financial markets and a level playing field amongst banks. Effectiveness relies on the commitment of all signatories to faithful and timely implementation in their jurisdictions. Postponing or watering down these last Basel III reforms in Europe puts at risk the broader trust European banks now enjoy, which will be needed in any future crisis, and has the potential to simply shift risks from one place to another.

As communicated in the joint EBA-ECB letter sent to the European Commission yesterday, we do not see any benefit in further delays. The implementation date has already been extended in light of the pandemic and thus the effects are unlikely to coincide with the timeframes for full implementation of the Basel III reforms.

The output floor

Let me now turn to the last point: the debate about the substantive effects of the output floor agreed upon by the Basel Committee.

This reform has a dual purpose: to reduce excessive risk-weight variability and to enhance comparability across banks. These two objectives are justified as there was evidence of significant and unwarranted variability in RWAs for similar exposure classes across banks. And although the output floor is the main contributor to the increase in minimum capital requirements, the impact seems manageable for European banks for several reasons.

First, it was calibrated at the lower end of the range that was discussed by the Basel Committee. It has been set at 72.5%, thus still offering considerable benefits to banks opting for advanced risk management approaches with internal models such as the internal ratings-based (IRB) approach for credit risk.

Second, the output floor is complementary to the supervisory work conducted at European level to improve internal models, such as the ECB’s targeted review of internal models (TRIM) and the EBA’s internal model repair programme. The TRIM project already strengthened how significant institutions comply with applicable regulatory requirements for internal models.[2] The impact of the TRIM exercise on RWAs was considerable, leading to an increase of €275 billion. Of course, this also reduces the gap between modelled and standardised RWAs, which reduces the impact of the output floor.

Third, the most recent EBA impact study for a sample of 99 European banks shows that: (i) due to the gradual phase-in, the impact of the output floor will be felt most in 2027 and 2028, allowing time for banks to respond; and (ii) among the 13 banks identified with an expected total capital shortfall in 2028, the output floor is the binding requirement for only six banks using internal models.

Aside from the impact, it is equally important to recognise that the output floor was a key element in the agreement of final Basel III reforms: implementing it is a necessity if the EU is to continue to be seen as a reliable upholder of international standards in the banking sector. Consistency with international standards also means implementing the output floor in such a way that there is only one amount of RWAs for each individual bank. As also highlighted in the joint EBA-ECB letter, we should avoid a “parallel stacking” approach whereby banks would be required to meet multiple requirements based on floored and unfloored RWAs. This would only increase the complexity of the framework, impair the comparability of capital ratios across banks and jurisdictions, make the banks’ internal risk management framework even more complex and have negative effects on transparency and thus investor visibility. It may also neutralise the intended objective of the output floor.

Some have argued that the output floor runs the risk of double-counting. We have communicated that the ECB is ready to mitigate any unintended effects arising from the accurate implementation of the output floor in Pillar 2. When setting our Pillar 2 capital requirements, we will avoid any double-counting of model risk and unwarranted regulatory drag from the recalculation of risk weights linked to the floor. If RWAs for the Pillar 1 capital requirements increase as a result of applying the floor, we will not let Pillar 2 requirements automatically rise in absolute value in the absence of a corresponding increase in the underlying risks.


In summary, I support the full implementation of the Basel III reforms. A faithful implementation is in the interest of all stakeholders. In the long run, we will reap considerable benefits in terms of resilience and continued and even enhanced credibility in the euro area banking sector. In the short term any potential conflict between the post-pandemic recovery and the Basel III reform is limited, of a transitory nature, and the one-year postponement of the implementation already mitigates the impact of the crisis. We should not postpone further or revisit the Basel Committee agreements in relation to this final leg of the reforms aimed at long-term structural change emanating from the great financial crisis. The output floor is key to the reform, and while it may have a considerable impact on a few European banks, there is no evidence of any outsized, across-the-board impact. Moreover, the phase-in allows time for a management response before the output floor becomes binding for some banks. Finally, we are committed to mitigating any unintended impact a well-implemented output floor may have on Pillar 2 requirements.

[1]See Budnik, K., Dimitrov, I., Groß, J., Lampe, M., and Volk, M. (2021), "Macroeconomic impact of Basel III finalisation on the euro area”, Macroprudential Bulletin, Issue 14, ECB, Frankfurt am Main, July.
[2]Meaning, for instance, that key parameters of credit risk models are calibrated towards long-term average default and loss rates, as well as the incorporation of downturn conditions in parameter estimates.

Speaking engagements

Media contacts