When fragmentation creates complexity

Speech by Patrick Montagner, Member of the Supervisory Board of the ECB, at the A&O Shearman SSM senior meeting

Kronberg im Taunus, 24 June 2026

Thank you for inviting me to speak here today.

Before I discuss overcomplexity in the banking system and what we can do it prevent it, I want to reframe the debate slightly. Because I believe there is a different question we should consider first: is complexity actually to blame for the issues facing the banking industry today? The short answer, in my opinion, is not entirely.

The problem of complexity has been a recent point of discussion for regulators in many other jurisdictions. In the United Kingdom, the Prudential Regulation Authority (PRA) has acknowledged that it has been applying the framework primarily designed for large, internationally active banks to smaller, domestically focused institutions, without sufficient differentiation. This has created a complex burden of compliance for small banks and building societies that yields, in the PRA’s own assessment, only marginal prudential benefit.^[1]

Since the beginning of 2025 officials from the Federal Reserve System in the United States have pointed to a similar concern around overlapping requirements. They noted that the same underlying risk can be captured simultaneously through risk-weighted asset frameworks, leverage ratios and stress-testing exercises. As a result, the risk may be recorded multiple times without necessarily providing deeper supervisory insights. More broadly, some US officials have highlighted the cumulative effect of post-crisis rulemaking as a primary driver of regulatory density, where layer upon layer of new requirements are applied without removing legacy structures.^[2]

These examples do not necessarily apply to the euro area, which operates within a significantly different institutional framework, and we also do not share all aspects of their diagnosis. However, they do point to real phenomena that are relevant to the global banking system, and they provide a useful starting point for this discussion.

It is also worth being clear about complexity. The same word can hold different meanings depending on the perspective of each stakeholder: the complexity of banks’ business models, the complexity of risks, the length and granularity of legal texts, the burden of supervisory processes, the accumulation of guidance, and the fragmentation created by national specificities. These are related, but they are not the same. A complex risk may require a sophisticated response. A fragmented process, by contrast, may simply require better design.

My own reading of the issue can be summarised in two key points. First, complexity in banking regulation is mostly a reflection of the reality of the institutions we supervise, the societies we serve and the risks that materialised when simpler frameworks proved inadequate. Second, when a genuine problem emerges, it often has less to do with complexity itself and more to do with fragmentation: a lack of integration in the banking system that generates redundancy, inconsistency and unnecessary burden without producing better prudential outcomes.

Complexity is largely a reflection of our reality

Why is banking regulation so complex? The simplest answer is that banking itself is complex. And, over the years, the environment that we operate in has become more complex in many ways; technologically, economically and geopolitically.

Consider the institutions we supervise here in the euro area. Large banking groups still perform the core functions of banking: they take deposits, make loans, manage liquidity and transform maturities. But the environment in which they perform these functions, and the channels through which risks can be transmitted, have changed profoundly. Today major banks operate across dozens of jurisdictions, and they run sophisticated trading books alongside retail deposit-taking, manage climate-related transition risks alongside traditional credit risk and process transactions at speeds that would have been unimaginable two decades ago. From T+1 settlements to digital asset tokenisation, the operational and risk profiles of these institutions bear no structural resemblance to those of the past. The same is broadly true of all European banking groups. We simply cannot apply the conceptual tools of the 2000s when we supervise institutions in 2026.

Another often overlooked dimension of this debate is that we need some complexity to accommodate the diversity of the European banking sector. Europe’s banking landscape spans from large, internationally systemic institutions to small cooperatives and savings banks serving local communities. And this diversity is extremely valuable for Europe. It means that institutions have different ownership models, different mandates, different risk profiles and contribute to the economy in different ways. But it falls on our banking regulation to accommodate this diversity, inevitably leading to more legal texts and long clarifications. From the outside, this can look like overcomplexity.

Given the diversity and complexity of the European banking sector, proportionality – adapting the nature and intensity of supervision to the specifics of the bank – is the right approach to take here, particularly in those countries where this diversity is more pronounced. Proportionality is already embedded in the European framework, including through much lighter reporting requirements for small and non-complex institutions (SNCIs). But there may be scope to go further, for example by tailoring the frequency or granularity of certain supervisory activities and by increasing the €5 billion threshold of the SNCI regime. But proportionality must apply to how banks are supervised, not the level of protection depositors can expect. A depositor in a smaller savings bank has the same right to the protection of their deposits as a depositor in a globally systemic institution. The principle that underpins the European framework – that savers here in Kronberg, or in Paris or Riga should have the same confidence in the prudential rules regardless of their bank’s size and location – is a fundamental promise of the Single Market. Proportionality must operate under that constraint.

It is also important to note that demand for complexity can come from the regulated sector itself. When facing supervisory discretion, the same institutions that call for high-level principles are also the first to request precise rules, detailed Q&As and explicit guidance that eliminates interpretive uncertainty. This is an entirely rational response, as excess discretion can lead to unpredictability and unpredictability can lead to legal and capital risk. But the consequence is a growing rulebook shaped in part by legitimate desire for clarity by those being regulated.

We should also be self-reflective. Supervisors can contribute to this dynamic when, in the name of transparency and consistency, we produce too many layers of guidance or anticipate too many cases. This is why ECB Banking Supervision is reviewing its supervisory guides, letters, reports and methodologies, with the aim of making them more concise, more consistent and easier to navigate, while drawing a clearer distinction between binding legal requirements and non-binding supervisory guidance.

All this requires a cultural change on both sides. Supervisors should focus on material risks and avoid turning risk-based supervision into a compliance exercise. Banks should take responsibility for applying the law to their own business models, rather than seeking guidance for every possible scenario. Simplification is therefore a shared project.

None of this makes complexity comfortable to live with. But it does suggest that complexity is not always the result of regulatory overreach. The more serious problem arises when different layers of rules, expectations and national frameworks do not fit together coherently and when fragmentation creates complexity.

The real challenge is not complexity, but fragmentation

The European regulatory framework faces two forms of fragmentation. The first is fragmentation within the prudential framework itself. The European prudential framework is fragmented across multiple levels of legislation, across jurisdictions that retain national options and discretions^[3] and across policy areas that have often been developed sequentially rather than in a coordinated manner. The result is a degree of structural incoherence that can increase regulatory and supervisory burdens for banks.

Since the banking union was created, EU Member States have sought to reduce unnecessary barriers to integration. Yet the fragmentation we observe today has a deeper structural dimension that extends beyond what prudential regulation alone can address. This is a second wider structural fragmentation.

The challenge of building a truly integrated cross-border banking system in Europe reflects the considerable diversity of national systems. Consumer protection frameworks, insolvency regimes, tax systems, labour regulations and banking product structures differ significantly across Member States. A mortgage portfolio, for example, cannot be easily integrated across borders without taking these differences into account. As a result, retail banking remains fundamentally local because the underlying legal and economic infrastructure is still national.

These differences reflect political and social choices made at the national level. Some of this diversity is understandable, but very often, additional national layers, divergent implementation or gold-plating also create complexity that is not clearly justified by local market specificities or by prudential benefits. These are not issues that fall within the mandate of a prudential supervisor.

The tools available to address this type of fragmentation are limited and specific. But we can act, and we are acting. On the regulatory side, further progress on the banking union – including, in due course, a European deposit insurance scheme – would do more to reduce structural fragmentation than any number of supervisory simplification measures. Where appropriate, greater reliance on directly applicable European rules can also help reduce national layering and improve consistency. But this is a legislative agenda rather than a supervisory one, and that distinction matters.

We are already doing what we can within our mandate. ECB Banking Supervision has launched a number of internal reforms. These include changes to the Supervisory Review and Evaluation Process (SREP); the “next-level supervision” project, which extends the SREP reform objectives to areas including decision-making processes, capital-related decisions and on-site inspections; and an initiative to promote a more unified supervisory culture.

Our SREP reform has introduced a multi-year, risk-focused approach that concentrates supervisory attention where it matters most. Supervisory decisions are now more concise and the number of corrective measures imposed has fallen significantly in recent years.

There is, however, an important paradox. When we seek to simplify supervisory processes – as we have done, for instance, through the fast-track process for significant risk transfers in securitisation – we often find that simplification requires more precise, rather than fewer, eligibility criteria. Making a process faster and more predictable means defining its conditions more clearly. In these cases, complex rules come at the price of simpler processes.

A similar point applies to capital requirements. While maintaining resilience, there may be scope to streamline parts of the capital framework, for example by simplifying the macroprudential buffer architecture.^[4] Banks experience capital requirements as a whole, with the overall capital demand reflecting Pillar 1, Pillar 2 and the macroprudential buffers set by national authorities with top-up powers by the ECB. Differences in national buffer frameworks can therefore contribute to variations across the banking union.^[5]

All this serves as a reminder that headline measures of regulatory complexity can be misleading. The alternative would be to remove rules and prior supervisory approvals in the name of simplification. That is clearly not the path the ECB has chosen. As a central bank and supervisor, we retain a strong institutional memory of past banking crises and of the costs of only identifying risks after they have materialised.

Conclusion

The case for simplification is real, and we are pursuing it. But the financial system we supervise is sophisticated, interconnected and rapidly evolving. The risks it generates – whether digital, climate-related or geopolitical – cannot be fully addressed through simplistic, rudimentary frameworks. Past crises, many of which occurred when regulation was underdeveloped, also remind us that a thinner rulebook is not, in itself, a guarantee of financial stability.

What we should aim for – and what ongoing reforms are seeking to achieve – is regulation that is proportionate, internally coherent and free from unnecessary redundancies that arise when frameworks are built layer upon layer without sufficient integration. Complexity that reflects the realities of the financial system is sometimes unavoidable. Complexity that reflects fragmentation and the accumulation of legislative compromises is what we can, and should, reduce.

In practice, this distinction is not always easy to draw. But drawing it carefully is more useful than calling, in general terms, for less regulation.

Thank you for your attention.