Ladies and gentlemen
Introduction
A few days from now will mark the third anniversary of the takeover of Credit Suisse by UBS. The events of March 2023 not only signalled the end of a major systemically important bank, but also the start of a fundamental debate on accountability, regulation, supervision and stability.
Just a few weeks after the crisis, I highlighted three specific statutory measures during a speech at the press conference that I believe are key if we are to draw the right lessons from the Credit Suisse case.
These include the statutory power for FINMA to impose fines in the event of serious violations of supervisory law, a clearer allocation of responsibility among senior management, and FINMA’s right to inform you – the public – about concluded proceedings involving serious rule violations. Since then, these demands have been raised many times: from the expert report in summer 2023, to the Federal Council’s TBTF report in spring 2024, to the report of the Parliamentary Investigation Commission (PInC) at the end of 2024 and finally, in summer 2025, in the Federal Council’s parameters for the upcoming legislative reforms.
In parallel to this, I have engaged in numerous discussions with politicians, media representatives, supervisors and citizens. Today, three years later, I would like to take stock and address various questions and aspects that keep coming up in the debate. I look forward to the opportunity to continue the discussion with you – the analysts of the Swiss Financial Analysts Association (SFAA) – after my presentation. The discussion is no longer purely technical. It is also political and decidedly controversial. And rightly so. Because what our financial centre will look like in 10 or 15 years’ time is decided today. Major legislative reforms are rare and affect us all.
It is generally accepted that there is no single measure that covers all aspects of reform and instead a package of measures is required. My speech today deliberately revisits the demands I made three years ago. I will explain in a moment why I am focusing on the three new legal powers I have already mentioned.
Before that, however, I would like to address the most widely discussed aspect at present: capital.
Basic considerations surrounding capital
Much has already been said on the issue of capital – in particular the capital backing of foreign subsidiaries. I would just like to make three basic points here.
First: In my opinion, the ongoing discussion about capital requirements does not always take into account the fact that “double leverage” – i.e. the partial use of debt to finance the equity of foreign subsidiaries – has been recognised internationally as a risk for over two decades. And the ongoing amendment to the law on “double leverage” is nothing new in Switzerland either. The same issue was discussed back in 2012 with very similar arguments. At that time, almost 15 years ago, Parliament essentially decided to continue to allow the partial funding of equity with debt. The proposal to abolish double leverage is therefore neither new nor typically Swiss and nor has it come about as a result of the CS crisis.
Second: The risks of loss associated with a systemically important bank can ultimately only be borne by two parties: Either by the bank itself, specifically by its owners and creditors, or by the taxpayer. The fundamental decision as to which of the two possible risk bearers should carry how many risks is not a technical question It cannot and should not be made by a specialist technical authority. Instead, it is a political decision – specifically a decision for the Federal Council and Parliament as elected representatives of the taxpayers.
Third: The same rules do not mean the same effects for every country. As a hypothetical example, let’s imagine two identical large banks with foreign subsidiaries. Both have the same business model, the same risk appetite and – to simplify matters – the “same” management. Let us further assume that both banks are subject to the same capital requirements, e.g. the same percentage of partial funding of the equity of their foreign subsidiaries with debt. The only difference between the two banks is their domicile. One is based in Switzerland, the other in the USA, for example.
From the bank’s point of view, the same rules apply. But the economic starting position is different: The USA has a large domestic market, but its foreign subsidiaries are small compared to the domestic market. Switzerland, on the other hand, has a small domestic market and the foreign subsidiaries are very large in comparison. The result: With the same capital rule, the Swiss taxpayer bears a higher risk than the American taxpayer in the event of losses at the foreign subsidiaries. The decisive factor is therefore not only the rules, but also the size of the domestic market relative to the foreign subsidiaries.
I will leave it at these three comments on capital. Besides capital, liquidity is also a central pillar of stability and therefore deserves attention – even if I am not going to go into this in depth today.
Prevention starts with the right incentives
Let me now explain why I emphasised the three legal powers I am about to discuss three years ago and why I am doing so again today. All three instruments serve the same purpose: prevention.
For years, financial stability has been understood as an equation with two variables: capital and liquidity. And rightly so. To put it simply: Capital absorbs losses. Liquidity absorbs stress. The global financial crisis of 2008 heightened awareness of the paramount importance of capital and liquidity. The lessons learned from this crisis still apply.
Both are necessary, but they are not sufficient. This was a lesson learned from CS. The numerous analyses and expert opinions on CS agree on one point. Let me paraphrase the report of the PInC: A combination of years of mismanagement, a questionable risk culture and massive misaligned incentives in the remuneration system were the main causes of Credit Suisse’s downfall. In addition, there were repeated scandals and loss events, which in combination led to an unstoppable loss of confidence on the financial markets. Rarely has there been such a broad consensus regarding a banking crisis that management played a major role in undermining market confidence and thus in the collapse of CS.
This is typical of banking crises: Before any problems emerge in liquidity and capital, there are shortcomings in a financial institution’s risk culture, business model and governance. To mitigate the risk of a banking crisis, preventive measures with statutory force are essential. The downfall of Credit Suisse demonstrated particularly clearly that the key factor is behaviour. But how can this be influenced?
“You can’t regulate decency” – “but you can learn it” – the role of incentives
Federal Councillor Karin Keller-Sutter said that “you can’t regulate decency” at the highly publicised press conference in March 2023. I fully agree with that. But as I did at the press conference on the Federal Council’s parameters in June 2025, I would like to add this: “Yes, you can’t regulate decency – but you can learn it.” Incentives play a central role in this “learning process”. This is not about technical details, but about the fundamental division of roles between the market and the state.
The strategy, business model and selection of management are the core elements of a bank. But strategic misjudgements on the part of the bank, management failings or losing the trust of clients and investors are not supervisory offences. Responsibility for action in the three core elements lies with the institution. Neither regulation nor supervision can or should assume the role of the owners (shareholders) or the board of directors. In particular, a supervisor’s “toolbox” is not there to impose a strategy on a bank or to restore the trust of clients. This is the task of the institution’s management and board of directors.
What then is the role of the state, specifically of regulation and supervision in these matters? A key task for our laws is to set the right incentives – the guiding principles, so to speak. Why is this important? Because the right incentives help prevent crises, while the wrong incentives make future crises more likely. If someone knows that they can profit from successes, but that others have to bear the responsibility in the event of a loss, then the temptation to take more and greater risks is big.
In particular, the incentives must be set in such a way that responsibility is borne symmetrically in both good and bad times. For example, if a business model is successful, the management and owners reap the rewards, because taking on risk and responsibility need to be rewarded. However, if a business model fails, the state cannot suddenly be held responsible by way of regulation or supervision. Figuratively speaking: The same responsibilities should apply in “rainy weather” and “sunshine”.
In the public debate, the opinion is sometimes voiced that regulation and supervision should generally “prevent” banking crises. But management failings, poor strategic decisions or a loss of market confidence cannot be completely prevented by supervision – and nor should that be the aim. Such fully comprehensive insurance would inappropriately blur the roles of the market and the state and thus set the wrong incentives.
This blurring of roles would also hardly be compatible with a market economy. Most people therefore agree on these basic tenets. The differences arise in the discussion of concrete measures. In my opinion, all three amendments to the law discussed here today are essential, as they set incentives in such a way that responsible behaviour is encouraged. This also means that serious rule violations can be communicated transparently and that these have tangible consequences so that rule violations are not indirectly rewarded. Our objective is therefore not to impose more rules, but to enable tougher consequences to be imposed for rule violations.
Fines
One much-discussed measure is whether FINMA should be authorised to impose a fine on a financial institution in the event of serious rule violations. This question must ultimately be answered by the legislator. However, it seems to me that too little attention is being paid to the main advantage of fines in this discussion. In my opinion, this lies in the clear signal they send out. There is hardly any other tool that conveys with such clarity – figuratively speaking “in big red letters” – “This institution has seriously violated its supervisory obligations.” It is not only clear, but also understandable for everyone. Because the message sent out by a fine does not only reach lawyers or specialists who know what enforcement is and know how to evaluate it. The clear signal sent out by the fine reaches everyone: clients, the investor community, employees and the general public. And the signal also reaches you, in particular, as analysts of the Swiss Financial Analysts Association (SFAA), thus aiding you in your interpretation of an institution’s rating.
The signal is not only clear and generally understandable, but the amount of the fine also makes it clear how serious the supervisory authority considers the misconduct to be.
The fine is hard to beat as a clear signal. This is probably why all – and I emphasise all – jurisdictions with which Switzerland generally compares itself use fines as an instrument of financial supervision (Figure 1, Fines).
But what if the legislator were to decide that FINMA should remain an international outlier in this matter and not be authorised to impose fines? It would be a mistake to believe that Swiss institutions would then not pay any fines. The only difference would be who decides on the fine and the amount – and where the money goes.
Let’s take Credit Suisse as an unfortunate example. In the case of Credit Suisse, payments for fines, settlements or damages between 2010 and 2022 amounted to around CHF 15 billion. These fines were decided abroad and paid abroad. However, it has been clear for all to see since the crisis weekend three years ago that Switzerland bears the risks of its systemically important banks. None of the billions in fines paid by CS ever flowed into Switzerland. Politicians will decide whether this should remain the case. I think it is legitimate to at least ask the question.
In the discussion about fines, the objection is often raised: “But CS has already paid billions in fines – and yet it has done no good.” I am not convinced by this argument. Firstly on a fundamental level. We accept fines in many areas of life. Despite parking fines, there are still parking offenders. Yet we do not hear people demanding that we abolish parking fines. I therefore ask you: A monetary fine is deemed to be ineffective in the money business of all things? I ask you another question: By this logic, shouldn’t we then also consider whether bonuses – in a sense the reverse of fines – are also ineffective? I will not go into this in detail and will leave the answer to you and the legislator.
There is an institutional reason that is even more important than these basic considerations. It makes a big difference whether or not a fine is issued by the same authority that also issues the licence. The authority that issues the licence and can therefore also withdraw it is responsible for primary supervision. Here, the signal sent out by a fine is much clearer and acts as a kind of yellow card.
A second element – besides the clear signal – also seems to me to have been neglected in the discussion so far and it may surprise you at first. A fine is a liberal instrument in many respects, one that promotes the market. Fines fulfil an important competitive function. If two competitors are active in the same market and one plays by the rules while the other does not, this distorts competition. A fine ensures that breaking the rules is not worthwhile. And vice versa: the fine protects those institutions that adhere to the agreed rules.
Fines also have another liberal advantage: The institution does not incur any preparation or implementation costs. An institution does not have to prepare specifically for a fine or create new structures. It is sufficient to comply with the existing rules.
Of course, the appropriate amount for a fine is also up for discussion: It seems important to us that a fine must have a deterrent effect in order to be effective. The amount of the fine should, in our opinion, therefore be linked not only to the severity of the offence but also to the institution’s performance. This ensures that fines are proportionate and no less effective for large banks than for small ones. The fine should be noticeable, but should not destabilise the bank financially. The fine is intended to have a preventive effect and thus be imposed at a time when the bank can still adjust its behaviour.
Technical arguments are sometimes put forward in the discussion. This is another objection we often hear: FINMA can already confiscate any profit that a supervised person or entity or a responsible person in a management position has made through a serious violation of the supervisory provisions – isn’t that enough? The following must be borne in mind here: The confiscation of profit presupposes that a profit has been realised in the first place. But here too, the CS case has painfully illustrated that serious breaches of supervisory law can also lead to high losses. In the case of the USD 5 billion loss in the “Archegos” case, for example, there was precisely zero profit that could have been recovered. By the way, the US and UK authorities fined CS USD 268.5 million and GBP 87 million respectively in this case.
Another objection to FINMA’s authority to impose fines relates to jurisdiction. The Swiss Bankers Association, for example, argues that FINMA, as the supervisory authority, has the power to investigate and should therefore not also be the authority to impose fines, as this would violate fundamental principles of the rule of law, such as the separation of powers. The fact is, however, that FINMA would have to justify any decision to impose a fine and substantiate it on the basis of the concluded investigation into the rule violation. And that each of these FINMA decisions could then be reviewed by the courts – the judiciary – in two instances.
Finally, I would like to discuss the crucial question of who should be fined. The institution or individuals? Sometimes people ask here: Is it fair for banks and therefore the owners to be penalised? Individuals are responsible for misconduct – shouldn’t they be punished?
The following should be borne in mind: Serious violations of supervisory law generally involve breaches of obligations incumbent on the institution. In other words, institutions should be organised in such a way that serious breaches of duty cannot occur. Ultimately, it is therefore the shareholders who bear all of the company’s opportunities and risks, whether they are directly individually responsible for them or not.
It is also important to distinguish between supervisory law and criminal law: Where criminal offences are committed by individuals, such as fraud or money laundering, criminal law also applies. This individual responsibility under criminal law exists independently of fines for an institution, and it could also be expanded if there is the political will to do so. Either way, however, the application of criminal law does not fall within FINMA’s remit – and we believe this should remain the case.
Last but not least, it is important to bear in mind that individuals who are responsible for a breach of supervisory law can – and should – also be held liable by the bank that has been fined. Because remuneration systems that allow employees to participate in the company’s success to a greater extent the higher their position in the hierarchy are also common, particularly in banks. However, holding people to account requires that responsibilities within the bank are bindingly defined and assigned. This leads me on seamlessly to the accountability regime.
Accountability regime
Much has been achieved in the international regulatory framework for financial institutions since the major financial crisis almost 20 years ago. The capital and liquidity requirements have been significantly tightened.
However, there is still a significant gap in one area: the personal accountability of managers. From comments sections to expert panels, it is regularly criticised that hardly any responsible managers have ever had to pay back bonuses or even face criminal prosecution.
Can this be changed? The applicable law requires a direct, causal link between a person’s behaviour and a specific breach of the rules in order for measures to be imposed on individuals. On the one hand, this requirement of a direct link is justified by the rule of law. On the other hand, providing proof of such a link is often a very high barrier in practice. This is particularly true for complex organisations – regardless of whether they are small or large. Here, it is extremely difficult to prove that a particular person was directly involved, was behind a problematic decision or deliberately ignored obvious risks. The financial supervisory authority must be able to prove precisely this individual connection. If it fails to do so, FINMA cannot sanction individuals.
This is where the accountability regime – known in international literature as the “senior managers regime” – comes in. The accountability regime is well known internationally and can be regarded as thoroughly tried and tested. As you can see, well-known financial centres already rely on a senior managers regime or are planning to introduce one (Figure 2, senior managers regimes). FINMA had already proposed the introduction of an accountability regime, which would allow the roles and responsibilities of managers to be more clearly defined, in 2022.
What does such a regime actually involve? The responsibilities within a bank must be set out in an overview document. On the one hand, this includes people who are already subject to a guarantee of irreproachable business conduct today. These are in particular members of the board of directors and executive board. But also – if they are not already included – other individuals at management level who can have a significant impact on the bank’s risk profile. Examples include the Head of Internal Audit or Compliance. A declaration of responsibility must be drawn up for each person in this extended group of members of corporate bodies, setting out their tasks and duties.
This in turn raises questions such as: Isn’t this creating a paper tiger that costs a lot and achieves little? It should be noted here that a clear allocation of responsibilities is already standard at well-managed companies – and should be achievable with reasonable effort for well-managed institutions. This existing allocation also allows the bank to take measures against offending managers itself in the event of violations, namely in the area of remuneration. However – and this should change under the senior managers regime – this allocation is not yet legally binding for banks.
In simplified terms, the introduction of the accountability regime essentially means that the existing organisation chart of an institution is made legally binding. Functions are assigned to clearly named people in a legally binding manner. These people bear an expressly defined responsibility for these functions. I emphasise: it makes it easier to prove a direct link between the person and the offence, but does not replace this proof. This is because the accountability regime does not lead to a reversal of the burden of proof, nor does it create an automatic liability mechanism.
Against the backdrop of an internationally recognised approach, the key parameters of such a Swiss accountability regime will be discussed. I am familiar with a version of this model from Asia, specifically from Hong Kong. The British model is also frequently cited. The two differ relatively strongly in their design. The Hong Kong model is principle-based and therefore comparatively lean; the UK model is rule-based and therefore comparatively broad.
In Switzerland, the regulatory tradition and supervisory practice of FINMA is known to be principle-based. A lean, principle-based accountability regime could therefore help to close the existing gap in personal responsibility – without jeopardising the principles of the rule of law.
FINMA is committed not only to the principle-based approach, but also to a proportional approach. Proportionality is in FINMA’s DNA. But being small is not a free pass. Rules of conduct such as the prevention of money laundering through to the assessment of appropriateness and suitability apply to small and large companies alike. However, it is appropriate and reasonable that small, well-capitalised and liquid institutions, for example, have fewer reporting obligations. The fact that we are the world’s first supervisory authority with a small banks regime shows how important this is to us. We have also had a small insurers regime since last year. Every year, small bank and small insurer symposia are held, at which FINMA and the small institutions share their experiences with the regime.
It is therefore clear that we are also in favour of a proportional accountability regime. It should be easy to integrate into existing structures with minimal bureaucracy, especially for smaller banks with short decision-making channels.
Transparency about enforcement
Unlike fines, it first needs to be explained to the general public what the term “enforcement” actually means. Enforcement is FINMA’s strongest weapon today. It is the compulsory implementation of supervisory law, if necessary by judicial means.
If there are indications of a serious violation of supervisory law, as a first step, the facts of the case are clarified in depth as part of an enforcement procedure. The second step is to restore compliance with the law. Enforcement takes place by way of a formal judicial procedure that is concluded with a ruling. Put simply, the latter encompasses both steps: describing the facts of the case and the measures to be taken by the institution.
How active is FINMA in investigating serious rule violations? On average over the past three years, FINMA employees have approached institutions two to three times a day to clarify the facts of a possible violation of supervisory law as a first step. That’s around 750 enforcement investigations per year. In the past year alone, this figure was even higher. Compared with international peers, that is a lot of investigations and a sign of a vigilant authority.
How forceful is FINMA? These investigations into the existence of a serious rule violation quickly have an effect at the vast majority of institutions. Following FINMA’s intervention, in over 90% of enforcement investigations full compliance with the law is restored within three months on average – including criminal charges. But 90% is not 100%. For around one in ten institutions, it takes longer – sometimes much longer (Figure 3, FINMA enforcement).
How many proceedings are concluded and how many of them are published? Under existing law, FINMA may not, in principle, report publicly on individual enforcement proceedings. The only exceptions are cases where there is a particular supervisory interest in doing so. To avoid destabilisation, this often makes sense. We publish figures on an anonymous basis and have consistently expanded the corresponding statistics in recent years.
However, we are very keen that we should be able to make our work more visible to the public in future in the event of serious rule violations – just as our supervisory colleagues in neighbouring countries are often already allowed to do. Their legislation is structured the other way round and generally permits the publication of concluded enforcement proceedings. There, non-publication is the exception. Here you can see the published enforcement cases as a proportion of all concluded enforcement cases in important financial centres. It varies between 95% and 100%. What is the publication rate in Switzerland? Only about 7%. You – the general public – hear on average just five times a year from FINMA about a concluded enforcement proceeding (Figure 4, Enforcement – publication rate).
It is not only the low permitted publication rate that is an issue, but also the time it takes to conclude a proceeding. There are several stages before a proceeding is concluded. Therefore, the conclusion of an enforcement case – and thus also the provision of information to you – can sometimes only take place years after the actual incident. This is especially the case when securing evidence is complicated, which is often the case with large and globally active financial institutions. So when the question is asked in response to scandalous headlines: “Where is FINMA?” it is important to bear in mind that FINMA is often not authorised to provide information on the steps it is currently taking in a specific case.
Three years ago, I called for greater transparency regarding FINMA’s concluded enforcement proceedings. The PInC also recommends strengthening the effectiveness of enforcement, along with increased transparency. The Federal Council has included this demand in the parameters, which I greatly welcome. Transparency helps to build confidence. And just like the other measures, it can be implemented cost-effectively. In addition, the market can react and have a disciplining effect, partly thanks to you as financial analysts. This is therefore also essentially a liberal concern.
Conclusion
Ladies and gentlemen. I have now come to the end of my speech: Three years after the fall of Credit Suisse, the question is not whether we have learnt any lessons. The crucial question is whether we are prepared to implement them.
Capital and liquidity are essential. Resolution planning is necessary for systemically important banks. But today, I have deliberately focused on prevention through corporate governance. Only clear accountability reduces the likelihood that tools such as liquidity, capital and the resolution plan will need to be utilised.
If accountability is blurred in good times, it will be shifted in bad times. If incentives are asymmetrical, the willingness to take risks will be excessive. And if rule violations have no tangible consequences, the rules will be circumvented. Not taking action in this area would effectively mean that the lessons of the Credit Suisse crisis have not been learnt.
Switzerland has a strong financial centre. It has efficient institutions and in-depth expertise. What it needs now is legal clarity that strengthens accountability, sets the right incentives and brings risk into line with consistency.
Banking is a private business. However, systemic risk is a public matter. Our task is to ensure that those who take risks also bear them – consistently, transparently and symmetrically.
Only then can market discipline have the intended effect according to a liberal understanding of the market. And only then can we honestly say that we have really learnt from the crisis.
Thank you for your interest and I look forward to your questions and the discussion.
