Banking regulators, supervisors and their watchers

Keynote speech by Claudia Buch, Chair of the Supervisory Board of the ECB, at the Watching European Financial Market Regulation and Supervision Conference, Goethe University Frankfurt

Frankfurt am Main, 15 September 2026

We all know about birdwatchers, and many of us know about central bank watchers. But how many of us know about watchers of regulation and supervision? Probably not many. This may seem surprising: savers, firms and taxpayers – they all care about sound banks.^[1]

So what is “watching regulation and supervision”, the theme of this conference, all about?

Ultimately, we are talking about accountability. As supervisors, we exercise public powers and carry out our tasks independently. Our independence requires accountability and transparency: the public must be able to assess how we use the responsibility we are given. Banks are the first to feel the effects of supervisory decisions. That’s why our decisions are open to administrative and judicial review.

But the effects of supervision and regulation on society are much broader, and the transmission channels are seldom straight.

How would banks have performed in the absence of supervisory measures? Is financial stability the result of good policy or good luck? How does the macroeconomic environment affect the performance of banks? What are the effects – and the potential unintended side effects – of regulation and supervision?

Answering these questions requires detailed data, good analytical frameworks and time. This partly explains why our watchers are a relatively rare species.

But we should not be content with the current state of play. We need to further develop the community of watchers.

Let me focus on three areas where progress can be made.

First, evaluations of regulation and supervision need to be comprehensive – assessing the direct effects on banks, but also the broader effects on the policy objectives that society cares about. The European Commission has recently published a report on the competitiveness of the European banking sector. The ensuing legislative process provides an opportunity to adopt that holistic perspective – looking at the implications of the prudential framework for growth, integration, financial stability and resilience. This is particularly important in the current environment characterised by heightened geopolitical risks and rapid structural changes in the financial system.^[2] Within our mandate to ensure the safety and soundness of banks, we are fully committed to contributing to such evaluations.

Second, we need a broad community of watchers. Policymakers, banks, the public, academics, analysts, journalists – they all have a legitimate interest in our work. Here, I see room for improvement. Often, banks are the most vocal in public debates. But we need a broader debate representing the depositors who entrust their savings to banks, the firms that are financed by banks and society at large which benefits from a functioning banking system but also bears the cost of crises. Academics have a special role and responsibility – they have the analytical tools to shed light on the complex functioning of the banking system and its welfare effects.

Third, we need to further develop the infrastructure that supports an informed debate. This infrastructure includes frameworks for policy evaluations, repositories of relevant studies, information platforms and data centres. It is essential to our accountability; it can promote independent evaluations of regulation and supervision. In the age of artificial intelligence (AI), we must harness advancements in terms of computing power and analytical capabilities to advance welfare.

The case for comprehensive evaluations of prudential regulation and supervision

There is a long tradition of watching central banks and monetary policy decisions among the media, academics and market analysts. Back in 1999, Ottmar Issing gave a speech entitled “The ECB and its watchers”.^[3] This marked the beginning of a series of conferences bringing together monetary policymakers and analysts.

While there is no direct equivalent to “ECB watchers” in the field of banking regulation and supervision, this conference can serve as a catalyst for change.

Monetary policy and supervisory decisions differ in terms of their objectives, instruments and communication channels. But these differences make assessing the effects – and potential side effects – of regulation and supervision no less important.^[4]

The supervisory policy cycle

As in monetary policy, a structured “policy cycle” – starting from the policy objective, the definition of intermediate targets and the choice of policy measures – can support policy evaluations. Let me give you an example – the setting of capital requirements.

The first step in the policy cycle is a clear objective. The objective of supervision is well defined in European legislation – to ensure the safety and soundness of banks, thereby contributing to financial stability.^[5]

The second step is identifying priorities and indicators that make progress towards that objective measurable. The safety and soundness of banks is shaped by sound capital, liquidity and governance and sustainable business models. These are the indicators we assess in our Supervisory Review and Evaluation Process (SREP) – some of them are quantifiable, some are more qualitative in nature.

Let me focus on capital as one example of how the policy cycle works in practice.

Bank capital is key. In essence, capital is a source of funding for banks. Unlike debt, it has the ability to absorb losses. It enables, rather than discourages, banks to lend and take risks. Regulatory requirements ensure that banks’ capital levels are commensurate with society’s risk preferences. In a crisis, risks should not be shifted away from bank managers and owners onto depositors or taxpayers.

In short: capital aligns incentives between banks and society.^[6]

Bank capital is also a valuable indicator as it can be tracked over time and across institutions. As recent ECB work shows, capital requirements increased after the global financial crisis, when resilience needed to be built.^[7] They have stabilised in recent years, with a shift from microprudential requirements to macroprudential buffers that can be released in times of stress.

This clear progress in terms of resilience is no coincidence. It is the result of improved regulation and the supervisory measures we take to ensure that a bank’s capital is adequate to cover the risks it takes – the third step in the policy cycle.

We set capital requirements based on a clear methodology. According to the Basel framework, Pillar 1 capital requirements are minimum requirements that address credit, operational and market risk.^[8] Pillar 2 capital requirements cover bank-specific risks that are not, or not sufficiently, covered by these minimum requirements.

This year, we are using a new, revised methodology for setting Pillar 2 capital requirements. This methodology is more risk-based, while leaving room for supervisory judgement and ensuring that there are no overlaps between Pillar 1 and Pillar 2 requirements.^[9]

Benchmarking is part of our methodological toolkit. Before informing banks of capital-related decisions, we ensure consistency across banks. A structured benchmarking process, conducted by an independent second-line function, identifies potential outliers and ensures that methodologies are used in a consistent way, while preserving judgement grounded in bank-specific risk profiles.^[10]

Evaluations and impact assessments

As a fourth step in our supervisory cycle, the effects of supervision can be evaluated ex post – at bank or aggregate level.

As regards the bank-level perspective, our analysis shows, for example, that supervisory judgement adds relevant information that automatic indicators cannot capture, while ensuring that supervisory discretion remains anchored in common methodologies.^[11]

But public interest in the effects of supervision is much wider, including the aggregate effects of overall capital requirements and resilience on macroeconomic outcomes. The European Commission’s report, for example, focuses on integration, innovation, growth and bank competitiveness. In the public discourse, distributional questions play an important role as well.

Generally, assessing the impact of supervision and regulation must isolate the effects of other factors. Macroeconomic conditions, market developments and management decisions affect outcomes. This requires sound evaluation frameworks and analytical rigour, a point to which I will return later.

Evaluations also need to consider the different layers of the capital stack. In addition to microprudential capital requirements, we set Pillar 2 guidance to ensure banks can absorb potential losses resulting from adverse macroeconomic scenarios. Unlike the legally binding Pillar 2 requirement, Pillar 2 guidance is a non-binding supervisory expectation based on banks’ resilience under adverse scenarios. As an integral part of the framework, the Single Resolution Board defines gone-concern capital requirements to ensure that the costs of bank distress don’t fall on taxpayers. National competent authorities set macroprudential requirements to safeguard financial stability.

All these requirements serve a purpose, but they also make the system rather complex.

Last year, the ECB Governing Council’s High-Level Task Force thus tabled proposals on how the capital stack can be simplified by combining macroprudential buffers into “releasable” and “non-releasable” types.^[12] Moreover, different requirements should be applied consistently across Europe.

Consistency and coordination is, in my view, what a “holistic” view of capital is all about – not the weakening or widening of policy mandates, or the setting of quantitative caps.

In sum, we need a societal debate about the broader effects of regulation and supervision, building on the best evidence and independent analyses.

As supervisors, we are strongly committed to contributing to these broader debates through our accountability channels, building a community of watchers and further developing the infrastructure for policy evaluation.

Building a community of watchers

Supervisors have a clear mandate to keep banks sound and safe. Within this mandate, they act as “delegated monitors” on behalf of depositors.^[13] Independence protects supervision from external influences. But this raises an important governance question: who monitors the monitor?

The key is accountability. It requires us to explain and justify how we exercise our powers – through several channels.

Generally, bank-specific supervisory decisions are subject to administrative and judicial review.^[14]

As supervisors, we are explicitly accountable to the European Parliament, which represents European citizens, as well as the Council of the European Union. We have regular exchanges of views with the relevant parliamentary committee and the Eurogroup, and we engage directly with national parliaments.

In addition, European banking supervision is subject to external scrutiny through the Commission’s periodic review of the SSM Regulation and through audits by the European Court of Auditors.^[15] And our work is regularly assessed under the International Monetary Fund’s Financial Sector Assessment Program.^[16]

Accountability requires transparency. Our Annual Report is the central vehicle for communicating how we carry out our supervisory tasks. The next report will provide information on the comprehensive reform of European supervision currently under way, with a view to becoming more efficient, effective and risk-based.^[17] Over the years, we have significantly enhanced our transparency through publications of methodologies, regular newsletters and blogs.

But a broad public debate about the effects of regulation and supervision goes beyond formal accountability arrangements. To effectively discuss and challenge our actions, this debate needs to reflect the views and perspectives of all stakeholders.

The Commission’s consultation on the competitiveness of the European banking sector is a case in point. It received 227 responses.^[18] Business associations accounted for 43% of the responses, individual banks for 16% and private-sector firms for 8%. Together, private-sector respondents made up around two-thirds of all responses. By comparison, public authorities accounted for 14% of responses, while academia, trade unions, NGOs and individual citizens together made up 19%.

These contributions focus on different aspects.^[19] At the risk of over-simplifying, there are two trends: banks tend to focus on aspects that have a direct impact on them and highlight the cost of capital requirements. NGOs and academics tend to emphasise the role played by the banking sector in achieving broader societal objectives, such as reducing climate-related risks, and they stress the importance of sufficient capital to support financial stability.

Each stakeholder offers a different and valuable perspective. But the uneven contribution of the various stakeholders shows the need to better balance the debate. The views of the public, savers and taxpayers are just as important as those of the industry. That’s why ECB Banking Supervision has stepped up its outreach to civil society across the banking union in the last few years. The Commission has several established outreach channels, such as the Financial Services User Group, representing consumers, retail investors and micro-enterprises.

Given that we are here at Goethe University, let me focus on what can be done to involve the academic community more.

More than 25 years ago, Ottmar Issing argued: “Although this may sound strange, we need to hear views even more when there is disagreement because this is when we learn. When views are not supportive of the ECB’s position but are well founded in good theory and obtain robust corroboration from the data, then again I think the ECB can only gain from a dialogue with academia.”^[20]

In this spirit, academics can bring analytical frameworks, rigour and independent assessments to the debate. They can focus on aspects which may otherwise receive insufficient attention.

Generally, our aim should be to build a stronger community of watchers, enhance the existing accountability channels and ensure that the debate is grounded in the best evidence. This requires further work to build a sound infrastructure.

Further developing the infrastructure for policy evaluation

To engage in birdwatching, you need good binoculars – and a lot of patience. Regulation and supervision-watching requires more than that: evaluation frameworks, repositories, access to data and information. This infrastructure has improved in recent years, but more in terms of investment and commitment is needed.^[21]

Frameworks for policy evaluation

Higher capital requirements are good for resilience – but could they come at a cost in terms of weaker growth or reduced lending? Improved regulations have made banks safer – but have risks shifted to other parts of the financial system so that, overall, the system is not necessarily more stable? Has European regulation had a negative impact on banks’ ability to successfully compete internationally? In short: what have been the consequences – both intended and unintended – of improved regulation and supervision?

These questions are at the core of current policy debates, but the answers are not straightforward. Answering them means going beyond descriptive statistics by disentangling the effects of the many factors that affect the performance of banks and the economy. Partial and general equilibrium effects need to be disentangled. Ultimately, identification of causal effects would be useful.

This is what evaluation frameworks are about – they help us learn systematically from experience. Both the Financial Stability Board (FSB) and the Basel Committee on Banking Supervision have such frameworks in place.^[22] At the European level, there are frameworks to assess the impact of EU legislation, including the European Commission’s Better Regulation Guidelines^[23] and the Regulatory Scrutiny Board which has, among other things, issued an opinion on EU supervisory reporting requirements.^[24]

These frameworks help manage the costs of policy evaluations. Not every aspect of the policy debate requires the same degree of scrutiny. Policy evaluations need to start from a discussion of the materiality of the issue at stake and the resources that are needed for evaluations. On this basis, evaluations can be prioritised.

Following this strategy, FSB evaluations have provided highly relevant insights into the effects of the post crisis reforms: the too-big-to-fail reforms, the implications of higher capital requirements on lending for small and mid-sized firms and for infrastructure finance, and the effects of reforms to securitisation.^[25]

The results of these evaluations show that the reforms have worked as intended, without material negative side effects on lending or growth.^[26] At the same time, they have provided insights into how the system can be further improved.

Similarly, the Basel Committee has a dedicated evaluation programme that complements the monitoring of the implementation of standards.^[27] This work confirms the results of the FSB’s evaluations: implementing the Basel III reforms increased banking sector resilience and reduced systemic risk, with no indication that the aggregate supply of credit to the economy had been impaired.^[28]

Importantly, these evaluation frameworks explicitly invite the views of stakeholders through public consultations, and they seek expertise from academics.

Further developing and using the existing frameworks will have significant benefits for upcoming debates on reforms.

Repositories and meta-analyses

Experience with evaluation frameworks shows that policy decisions need a comprehensive empirical foundation. Single studies are not sufficient; results need to be replicated over time and across countries.

Repositories and meta-analyses can help by pooling dispersed findings. They are routinely used in other fields, such as medicine, where identifying the right treatment and separating intended and unintended effects is indeed critical.^[29]

In the field of financial regulation, the Bank for International Settlements maintains a repository which can form the nucleus for additional work. The Financial Regulation Assessment: Meta Exercise (FRAME) contains studies on the effects of capital and liquidity regulation as well as the too-big-to-fail reforms.^[30]

These studies show that the impact of capital requirements needs to be assessed within the context of the cyclical environment and banks’ starting conditions.

Banks entering a downturn in a stronger capital position tend to perform better and maintain lending more effectively than banks starting in a weaker capital position. Weakly-capitalised banks are more likely to restrict credit when losses materialise.^[31] At the system level, stronger banks may then use the opportunity to gain market share and increase lending.

Over the longer term, higher capital requirements do not reduce lending.^[32] But when individual banks are already close to their minimum requirements, have weak profitability or face high costs to raise equity, increasing capital requirements can have a negative short-term impact on credit supply.^[33]

Conversely, reducing capital requirements does not necessarily generate additional lending. The supporting factors that Europe has introduced as (temporary) deviations from the Basel framework are a case in point. Their purpose is to ensure that lending to small and medium-sized enterprises (SMEs) and infrastructure is not unduly disincentivised by bank capital regulation. These deviations indeed lead to lower capital requirements for the affected exposures. But there is no evidence that they have a relevant impact on lending. According to the European Banking Authority (EBA), the SME supporting factor in particular did not stimulate lending to small companies relative to larger ones. For the infrastructure supporting factor, the EBA found that the data are insufficiently detailed to identify impacts on lending or risk.^[34]

These findings support the logic of building releasable buffers when economic and banking sector conditions are favourable, which banks can then use in bad times. But the benefit of releasing buffers during a crisis does not imply that permanent reductions in capital requirements during normal times will stimulate credit or growth.

Overall, these studies are highly informative for the current policy debate. There is also a strong case for further developing repositories, potentially using artificial intelligence to identify relevant studies and extract information. I see two main areas for improvement.

The first is to add more recent studies. Much of the analytical work on the impact of capital on bank lending was done around the great financial crisis. Adding more recent studies to FRAME would therefore be of great value.

The second is to broaden the perspective and add studies on other outcome variables, such as bank performance and the cost of capital. Recent work shows that, broadly speaking, capital requirements for banks are comparable internationally and not a relevant driver of performance.^[35] Rather, banks’ ability to compete successfully is driven by structural factors, such as the adoption of technology and the ability to achieve scale and operating efficiency.^[36]

Moreover, better-capitalised banks tend to face lower costs of funding, regarding both debt and equity.^[37] This shows a positive trade-off: equity capital typically carries a higher required return than debt. But higher capitalisation also makes a bank less risky, reducing required returns. Effective and proportionate regulation can thus lower the cost of capital, while weak and ineffective regulation tends to raise it.^[38]

Information platforms and data centres

Without sufficient data and information as an input, evaluation frameworks remain rather empty.

There is little shortage of aggregate information about the banking sector. ECB Banking Supervision, for example, publishes a wide range of supervisory information and data, mostly at an aggregate level in its ECB Data Portal.^[39] In 2023 quarterly data on less significant institutions were added to those on significant institutions. Over time, we have also added more detailed breakdowns including the cost of risk, funding ratios, net interest margins, Stage 2 exposures, non-performing loans and liquidity buffers. During the pandemic, we published specific information on loans subject to COVID-19-related measures.

Data for policy evaluations, however, need to be more granular. But, traditionally, access to granular, bank-level information has been more limited for legal reasons, in particular when it comes to comparable datasets across countries. One way to address this limitation is to run similar empirical studies across different datasets, thus bringing together the output, not the data. The International Banking Research Network has successfully implemented this approach for more than a decade now.^[40]

But access to granular banking data is now improving, following the lead taken by several national central banks.^[41]

This year, the European Banking Authority (EBA) launched its Pillar 3 data hub, which provides access to prudential disclosures from individual banks.^[42]

In addition, the EBA is developing an EU-wide public repository of European and national supervisory data requests, which is expected to become operational in early 2027. The repository will increase transparency about the data requested from banks.^[43]

Last year, the ECB started a pilot project giving external researchers access to granular, statistical, bank-level data. Technically, access is organised via research data centres within the Eurosystem, showing the benefits of operating a decentralised system. The pilot provides anonymised balance sheet information for around 2,000 banks and bank-level interest rate information on loans and deposits for around 300 banks for the period from 2007 to 2023. We also publish selected stress test results at bank level.^[44]

These are positive developments. Access to granular data is provided while maintaining the safeguards required for statistical confidentiality. But we need to do more in order to enhance transparency and data access in a legally sound way.

In addition to data, information about relevant legislation and supervisory practices is needed. The EBA’s Interactive Single Rulebook provides a searchable compendium linking EU banking legislation.^[45] Moreover, we are reviewing our supervisory guidance and, as part of this, we will make our guides more accessible and enhance the consistency of the language used.^[46]

Not least, international work can help improve the evidence base for comparing capital requirements across jurisdictions. Recent work by the Financial Stability Institute, for example, brings together harmonised information on capital requirements for global systemically important banks.^[47] This helps make the capital stack more transparent and supports a more informed debate on international comparability.

Towards strengthening the community of watchers

Banks affect everyday life. If they perform well, the payments system runs smoothly, deposits are safe, and firms and households receive loans. But if banks are under stress and insufficiently resilient, society bears the cost in terms of disrupted services, weaker growth, higher unemployment or even outright fiscal costs.

We therefore need to have an informed societal debate about the effects of regulation and supervision, based on evidence that can withstand serious scrutiny. This debate is needed to maintain resilience in the financial system in an environment of heightened geopolitical risks and rapid structural change due to the digitalisation of financial services.

The stakes are high. The current debate about the role banks play in promoting growth, stability and welfare should lead to changes in regulation that can make the system stronger. But there is also a non-negligible risk that the balance might be tilted towards the perceived short-term benefits of weaker standards at the expense of higher long-term risks.

This is why a broad community of watchers matters: policymakers, banks, researchers, analysts, journalists, civil society and citizens. Each group brings valuable input. Academics have a special role to play as they bring analytical rigour.

That is how birdwatching developed. Individual observations become more valuable when they are consistently recorded and shared among a wide audience. This can reveal patterns that no single observer can see on their own, helping everyone learn more about how the entire ecosystem works.

As European banking supervisors, we are strongly committed to further developing a community of watchers. We need a robust infrastructure for information sharing and rigorous impact assessments, including evaluation frameworks, we need repositories and we need transparent access to data and information. This is the foundation of our accountability.