Why I believe Europe must stop counting bank capital twice Guest contribution in The Banker online

The European Commission is right to recently argue the European financial system remains too fragmented along national borders. In practice, cross-border financial flows in the EU remain subdued and financial integration is still far from complete.

Some argue that differences in national regulations create friction for cross-border banking. But not all such differences are undesirable. The EU’s multilevel supervisory architecture deliberately combines European and national perspectives. Within the Single Supervisory Mechanism, significant banks are directly supervised by the ECB, while smaller, less significant institutions remain under national supervision, with ECB oversight.

Macroprudential policy follows a similar logic of shared responsibilities: primary responsibility for domestic financial stability rests with national authorities, while the ECB takes a cross-border perspective and can impose stricter measures where necessary.

There are good reasons for this. Europe’s financial systems do not always face the same risks at the same time. A housing boom in one country may coincide with a downturn in another, for example. Greater financial integration will not eliminate such differences.

Importance of national discretion

National authorities are best placed to address the systemic risks in their jurisdictions. Greater convergence around common standards and best practices at the European level is both possible and desirable. But proposals to limit national discretion over macroprudential policy should be approached with caution.

But keeping macroprudential policy in national hands matters only if the tools available to them work. This is where the Commission’s plans should be more ambitious.

Among the most important tools available to national authorities are macroprudential capital buffers. They require banks to hold additional capital against systemic risks. When a crisis hits, authorities can release the buffers – lowering requirements and giving banks more room to absorb losses and continue lending rather than deleverage and shed assets.

In practice, however, banks’ shock absorbers do not always work. A national macroprudential authority may release a buffer, only for another requirement, such as from the resolution framework, to become immediately binding. This can happen because the same capital used for buffers can also count towards other requirements. Releasing the buffer in a crisis may therefore create little or no additional usable headroom for banks, limiting the effectiveness of a central macroprudential tool.

The issue of limited buffer usability is material for European banks, as reports by the ECB and the European Systemic Risk Board have shown, and can weaken Europe’s resilience in a financial crisis. One solution, which I have repeatedly advocated, would be to prevent capital held as macroprudential buffers from also being used to meet other requirements, such as resolution requirements.

Under the EU’s current framework, the minimum requirement for own funds and eligible liabilities (MREL) is designed to ensure that banks have sufficient resources to absorb losses and, where necessary, be recapitalised in resolution. However, the same capital can count towards both MREL and macroprudential buffers. This overlap can limit the usability of those buffers in a crisis.

These functions should therefore be separated. Capital held for macroprudential buffers should not also count towards MREL. Releasing a buffer would then create genuine capital headroom that banks could use to absorb losses and support lending in a crisis.

Legislative change needed

I recognise that removing this double-counting of capital for buffers and resolution requirements would require changes to EU legislation. In some cases, it could affect banks’ capital needs, which may require a recalibration of the broader set of requirements. 

But the objective of this reform would not be to increase overall capital requirements, but to simplify and organise the framework more coherently, so that its different elements can perform the functions for which they were designed – including ensuring that buffers are usable when needed. Such a reform may take time and require close collaboration between microprudential and macroprudential supervisors and resolution authorities, but the ambition is warranted.

In the meantime, the EU should implement the Basel output floor as planned. The output floor limits how far banks using internal models can reduce their risk-weighted assets relative to the standardised approach. By placing a lower bound on model-based risk weights, the output floor can reduce the extent to which other parallel unweighted requirements become binding when macroprudential buffers are released. This increases the likelihood that releasing a buffer actually creates usable capital headroom, alleviating the problem of limited buffer usability but not solving it.

Deeper financial integration should not come at the expense of national macroprudential flexibility. A more integrated financial system will only be resilient if authorities can act when risks emerge – and if the tools they use actually work effectively.