Interview with Frank Elderson, Member of the Executive Board of the ECB and Vice-Chair of the Supervisory Board of the ECB, conducted by Aloysius Widmann on 14 October 2026
20 October 2026
Almost without exception, banks complain about the high cost of regulation. Do they have a point? Or are these costs essential for the stability of the banking system?
We shouldn’t forget the financial crisis of 2008. The costs were enormous, not only for banks, but also for households, firms, taxpayers and society as a whole. Those costs far outweigh the cost of regulation. The regulatory reforms introduced after the crisis have made European banks far more resilient. Banks today are well capitalised, have high liquidity levels and are more profitable, on average, than for many years.
But wouldn’t it be possible to simplify supervision without putting the stability of the sector at risk?
It’s true that regulation and the supervisory framework have become more complex over time. One of the reasons for this is that the banks themselves differ greatly and operate in an increasingly complex environment. The risks have also become more complex, ranging from geopolitical tensions, cyber risk and climate and nature risks through to artificial intelligence.
But complexity is not an end in itself. We do of course ask ourselves constantly whether we could still achieve our objectives as supervisors with simpler rules and regulations. And we’ve already taken a whole range of measures. We’ve reduced reporting obligations and accelerated processes. But it’s important to remember that simplification is not the same as deregulation. And it is resilience that is the foundation of sustainable growth and competitiveness.
Regulation means costs, and costs can be a competitive disadvantage. What do you see as the biggest obstacles to international competitiveness for European banks?
The biggest obstacle to European competitiveness is fragmentation. Although we have made progress, Europe still doesn’t have a truly single, integrated banking market. Cross-border banking activities remain limited. Banks are not able to fully benefit from economies of scale, greater diversification or more efficient allocation of capital and liquidity across the European economy. This fragmentation leads to higher costs, lower efficiency and fewer opportunities to achieve the kind of scale that is needed to compete internationally. This doesn’t just affect the banking sector but many other sectors besides. We talk about the Single Market, but the reality is the market is not yet truly single.
If fragmentation is the diagnosis, integration is the remedy – not deregulation. We need to complete the Single Market, the savings and investments union, and the banking union, including establishing a European deposit insurance scheme.
Simplification and integration are two sides of the same coin, which is the competitiveness agenda. We need both and we need both at the same time. This is why it is crucial that the European Commission’s banking reform package keeps both sides together.
How much would a fully integrated banking market actually encourage cross-border lending? Local banks know their customers best. And competition is already really strong in countries like Austria.
I haven’t got a crystal ball. What is clear, though, is that a single jurisdiction with several hundred million customers would lead to more competition, better products and more risk sharing. Banks would also be in a position to fund future-oriented investments; the challenges associated with the climate and nature crises, the energy transition and demographics are enormous.
Today, differences in legal frameworks, insolvency regimes and consumer protection rules still make cross-border banking unnecessarily complex and costly. No one is forcing us Europeans to keep these obstacles in the market. We’ve built them ourselves – we can, and should, also remove them and do so urgently. And with the enormous investment challenges in front of us, we’ve no time to lose.
Other countries could relieve the burden on their banks by implementing Basel III more loosely. Wouldn’t this cause Europe to become less competitive?
Banking is a global business. Capital flows across frontiers, banks are interrelated, and risks don’t normally come to a halt at national borders. That’s why international standards are so important. We want banks to compete on how efficient and innovative they are and how well they manage their risks, and not on who operates under the weakest rules. Thanks to the Basel standards, this is what’s happening, because we’ve created a broadly consistent international framework. The Basel standards create a common minimum benchmark for banking regulation and prudential supervision. Of course there is also space for national variations, where these are justified, carefully targeted and supported by evidence. But in a global financial system where shocks can propagate across borders, policymakers share a responsibility to observe minimum standards. Simply lowering the standards does not cause the risks to disappear. Supervisors will still need to assess the risks and ensure banks address them appropriately. Loosening Pillar 1 requirements could therefore simply shift the burden from regulation to supervision.
You’re calling for supervision to be more proportionate, meaning it takes more account of the size and complexity of the institutions that are supervised…
Proportionality is not about weakening supervision. The aim is to tailor the supervision better to the banks, using criteria such as size, complexity and risk profile. This is already embedded in the legal framework. For instance, there is the category of small and non-complex banks that are subject to significantly lighter reporting requirements. They only have to report up to 30% of the data that large banks are required to report.
But there is room to do more on proportionality. And I staunchly support doing just that. Right now, the threshold for whether a bank is considered small and non-complex is €5 billion in total assets. Raising that threshold to €10 billion would enable a further 150 banks or so to come under the simplified regime. Another option would be to lengthen the periods between regulatory reviews, where risk profiles allow. And for stress tests, we could make greater use of data that we already have to hand. All these changes would reduce the workload for smaller and less complex banks.
How far can such simplifications go?
No further than the risk. Proportionality can only go as far as a bank’s risk profile allows, and only as far as it does not impair our ability to safeguard financial stability and ensure banks are safe and sound. There should be no change in the core prudential standards. Banks still need to have sound corporate governance, effective risk management, adequate capital and liquidity and resilience in the face of new risks. Not only would smarter and more proportionate supervision lessen the burden on banks, it would also free up resources for us supervisors to concentrate on the most important risks.
What risks does artificial intelligence raise for banks?
AI has enormous potential for banks. It can improve risk management, facilitate the processing of information, reinforce fraud detection and raise efficiency. But every transformative technology also creates risks.
One area of concern is cyber risk. Advanced AI models can discover and exploit vulnerabilities more quickly. This is why, back in July, we asked the banks to reassess their cybersecurity measures and to review their IT budgets and staffing levels. They are expected to submit concrete action plans for dealing with AI-related cyber risks to us by the end of October.
Banks should not become overly dependent on AI models either. The technology has many advantages – helping to improve credit assessments, for instance. But banks need to watch out for biases and blind spots; they can’t simply leave decisions to the algorithm. A third issue is data quality. AI is only as good as the data it is trained on. And fourthly, operational dependencies on cloud providers, for instance, are emerging. The fact that there only so many such providers is giving rise to considerable concentration risk.
The AI investment boom in the United States is increasingly funded via private credit – loans extended outside of the banking sector – and corporate bonds. It remains to be seen whether the AI firms can actually generate the predicted profits that they use to justify the investment. To what extent does this create risks for European banks?
The key issue here is the interconnectedness between banks and non-banks. We’re taking this topic very seriously and want to see that the banks are also well aware of these risks.
A more diverse financial system is a good thing. Europe needs huge investments to address its strategic priorities, and market-based finance can play a role there alongside banks. Certain other players are better able to bear specific risks than banks. But if lending migrates out of the banking system, it does not automatically mean that all the associated risks migrate with it. Those risks can feed back into the banking system, because of the interconnections between banks and non-banks. It may be the case, for instance, that a bank not only lends money to a private credit provider but at the same time funds firms to which this provider also extends loans. This can give rise to risk concentrations that are not evident at first glance but that come to light when something goes wrong. That is why banks need to take a very close look at these linkages and make sure they can manage them appropriately.
