Thedéen: Trends in capital rules for banks

 

What are the capital requirements for and what is macroprudential policy?

But I thought I’d start by explaining why banks need to hold equity capital and why the government needs to impose capital requirements on banks. Banks play an important role in society and they need to be able to fulfil their basic financial functions of supplying payments, allocating capital efficiently and ensuring that risks can be managed effectively. It is therefore important to ensure that the banks can cope with losses, maintain stability, protect depositors, have the right incentives and avoid financial crises. At the same time, it is difficult to determine what the right level of capital requirements should be. Excessive capital requirements can limit the banks’ ability to finance investment and contribute to growth. On the other hand, capital requirements that are too low can increase the risk of banks failing and financial crises arising, often leading to large welfare losses. Finding optimal capital requirements is a delicate task. Many studies have attempted to estimate this, but the results are uncertain and only indicative.[1] 

Macroprudential supervision, for its part, aims to manage systemic risks in the financial sector and thereby reduce the risks to the economy in general. It has two dimensions. The structural dimension focuses on more permanent systemic risks that often result from dense interconnections among financial actors or high concentration and distorted incentives. Some banks may also be so important to the economy that we cannot let them fail. They are “too-big-to-fail”. Such banks therefore need to hold more capital as the entire economy could suffer damage if one of them fails. They therefore have to withstand greater losses than other banks. This is why there are capital buffers for systemically important banks, which are intended to increase their resilience and reduce externalities.[2] 

The time-varying dimension focuses on how vulnerabilities build up in good times, and then amplify crises if they materialise in bad times. The global Basel standards therefore introduced the countercyclical capital buffer to counteract this. Banks should build resilience in good times by building up an additional capital buffer and then being able to use the buffer in bad times, including to cover losses and maintain lending to creditworthy households and firms.[3] 

Systemic risks and macroprudential policy are of great interest to the Riksbank as we have a responsibility for financial stability. We therefore look forward to taking over responsibility for the Swedish countercyclical capital buffer. There are at least two reasons why the responsibility for the countercyclical buffer fits well with the Riksbank's responsibility. First, we already thoroughly analyse the real economy and the performance of different sectors in the context of monetary policy. They are useful for analysing credit market developments and related systemic risks. Second, the countercyclical capital buffer has become more of a crisis measure than was probably originally intended. The Riksbank has a clear crisis mandate and can now more easily and quickly coordinate decisions on the countercyclical buffer with other crisis measures. During the coronavirus pandemic, we saw that Finansinspektionen’s lowering of the countercyclical buffer also took place largely at the same time as the Riksbank introduced various crisis measures. 

Key issues for the countercyclical capital buffer 

The countercyclical capital buffer raises a number of interesting policy questions. I intend to address three of them today.

First, the way the buffer is used has changed over time. Originally, the buffer was to be built up as risks in the financial system increased. The idea is thus that the buffer should manage systemic risks that build up within the financial system. These include, for example, the risks arising from rapidly increasing debt levels in different sectors. However, the tool also proved to be useful for mitigating shocks from disturbances outside the system. The outbreak of the coronavirus pandemic is a clear example. Uncertainty was monumental at the time and there was a high risk that credit supply would tighten. In such a situation, a decision to reduce the countercyclical capital buffer may be even more important. This increases the distance to the requirements imposed by the authorities and may therefore put the banks in a better position to continue lending to creditworthy households and companies. In Sweden, Finansinspektionen lowered the buffer from 2.5 per cent to 0 per cent in March 2020. However, in many other countries, the authorities could not do so because the buffer was zero when the crisis hit. So in practice, Sweden had better tools to use than some countries as we had built up the countercyclical buffer in line with the rapid increase in lending. Countries that had not activated the countercyclical buffer had a more limited toolbox. Therefore, several countries in the EU have now introduced a so-called positive neutral level.[4] This means that the buffer has a target level even if there are no clear cyclical systemic risks. In Sweden, the Finansinspektionen has had a positive neutral level of 2 per cent for some time. The Riksbank supports this and intends to continue in this direction. It is also important that the decision-making on the countercyclical buffer is carried out with a certain degree of independence, especially if there is a positive neutral level. It is often politically easy to lower the buffer but not as opportune – and thus perhaps more difficult in some countries – to raise it again. 

Secondly, as authorities, we need to see if we can increase the usability of buffers more generally. This applies not only to the countercyclical buffer but to all buffers. The buffers are used both if a bank makes losses and if it continues to provide loans that require capital cover. They are intended to be used in stress, but international experience shows that the banks are often reluctant to do so. This view is also borne out by my discussions with bank management teams, most of whom in fact regard all capital requirements – including the buffers – as mandatory, and they will go to great lengths to ensure they do not breach them. One reason for this is that the use of buffers, as well as imposing restrictions on the bank, can be perceived as a weakness. Some banks prefer instead to hold back or even reduce their lending. If many banks do this at the same time, the supply of credit is reduced, which can have consequences for the real economy and exacerbate what is known as a credit crunch. In practice, only what are commonly referred to as management buffers comprise capital that is really useful for the banks to reduce the risk of a credit crunch. This is the capital that banks voluntarily hold in addition to the regulatory capital requirements. 

One way of making the buffers more usable for the banks is to give the countercyclical capital buffer a larger role and instead possibly reduce more structural buffers such as the systemic risk buffer. The uniqueness of the countercyclical buffer is that it can convert de facto mandatory capital requirements into usable capital. This buffer is therefore “releasable”. When it is released, the distance to the capital requirements increases; the bank's room for manoeuvre increases and the risk of an unnecessary credit crunch decreases. This means that the countercyclical buffer can more effectively counteract a credit crunch than other more structural buffers. A good example, as I understand from discussions with a number of Swedish banks, is the reduction in the countercyclical buffer at the start of the coronavirus pandemic. This played a positive role in encouraging the banks to continue lending at a time when there was a great deal of uncertainty. And it mattered that the distance to the mandatory capital requirements increased. One idea for the future could therefore be to allow for greater flexibility in the buffer requirements, so as to make them more releasable. Of course, this can be done in various ways; perhaps the buffers could be combined and given the same characteristics as the countercyclical buffer. In my view, it would be desirable to move towards designing capital requirements in such a way that they consist of more releasable and thus more usable capital. If designed correctly, this should promote both macroeconomic and financial stability. 

Thirdly, we need to review the reciprocity rules within the EU, i.e. the regulations stipulating that, in principle, the same rules should apply to all banks in a given market – regardless of which country they come from. This is especially true if we are to increase the importance of the countercyclical buffer. Today, the mandatory reciprocity of the countercyclical buffer is limited to 2.5 per cent. This means that if we were to theoretically raise the Swedish countercyclical capital buffer to, for example 3.5 per cent, it would fully affect the Swedish banks' exposures in Sweden, but the foreign banks' exposures in Sweden would only be affected by a buffer of 2.5 per cent, unless other countries choose to voluntarily mirror the Swedish increase for their banks in Sweden. When we have a positive neutral level of 2 per cent, this limits the scope for action. It may be difficult for us to raise the level above 2.5 per cent, as it could create problems regarding competition between Swedish and foreign banks in Sweden. The Riksbank has therefore argued for a number of years in favour of the EU removing the ceiling for mandatory reciprocity or raising it substantially. As more countries adopt a positive neutral level, this is likely to become a key policy issue, in which we hope more countries will support our line.

Simpler rules but preserved resilience

Over the past year, the international discussion has focussed on the fact that capital requirements for banks are complex – too complex. 

I hinted at an example of this complexity earlier. Allow me to go into a little more detail here. The Swedish countercyclical buffer is currently set at 2 per cent and is applied to all exposures to counterparties in Sweden. A Swedish bank with lending in other countries must also take into account the countercyclical buffer in those countries. In this way, a bank-specific countercyclical buffer can be calculated. On 31 December 2025, the bank-specific requirements were 2.04 per cent for Handelsbanken, 1.63 per cent for SEB and 1.77 per cent for Swedbank. For these three banks, the Swedish part of the buffer alone totals more than EUR 33 billion. This is not a negligible sum. However, if the Riksbank were to lower the countercyclical buffer tomorrow, it is not certain that the banks’ capital scope to expand lending would increase by the same amount. This is because the banks need to fulfil a number of different, parallel requirements. Larger banks need to fulfil as many as eight parallel requirements, some of which are linked to one another. If one requirement is lowered, another requirement may become binding, which means that the capital space does not increase as much as one might imagine at the start. For example, if the risk-based capital requirements are reduced, the non-based capital requirements or one of the MREL requirements may become binding instead. The system is difficult to understand, even for the authorities. 

There are several reasons why regulatory frameworks are complex. The banks’ operations are complex and the risks cannot be measured in just one dimension. At the same time, the regulatory framework should suit many types of banks operating in different conditions and markets. Moreover, it has evolved gradually over the years, building on top of existing regulations. Much of the regulatory framework is based on international agreements. Such international standards are essential because risks are easily spread across borders. This became very clear in the global financial crisis. However, the standards should be negotiated and applicable in many different jurisdictions with different traditions and banking systems. When standards are introduced in different countries, for example within the EU, the rules often become more complex. Part of the complexity stems from the fact that the rules are tailored to the specific risks in each country, which can help to build resilience. However, some of these issues also arise because countries and their banking systems wish to draw up rules that benefit their own operations. All in all, the result is a rather complex system.  

I therefore share the view that the current regulatory framework for banks is too complex and needs to be simplified. The discussion has started but will continue for a long time. Sam Woods, who is responsible for banking supervision at the Bank of England, gave a speech on this topic back in 2022.[5] In December last year, the ECB published a report containing a number of recommendations for simplification.[6] Borio et al have also published a report with a number of ideas.[7] The Draghi report sets out a series of proposals to strengthen European competition and argues that an important part of this is to simplify banking regulations.[8] In February, the European Commission launched a public consultation in preparation for the publication in the third quarter of this year of a report on how to increase competition and simplify regulations at EU level.[9] 

Without pre-empting the forthcoming discussions, I would like to share some general principles and personal reflections on this as Governor of the Riksbank. However, these should not be regarded as the views of the Basel Committee. 

One starting point is that it is good if the regulations can be simplified. Complex regulations entail costs. This makes it harder for banks, analysts and authorities to understand and act appropriately, as the consequences can be difficult to grasp. If the regulatory framework within the EU was simplified, it could also create the conditions for better competition between the banks within the Union. This raises the question of how harmonised the regulatory framework should be within the Union and what possibilities Member States should have to introduce their own stricter regulations to take account of local conditions and risks. Part of the complexity of the current regulatory framework is due, as I mentioned earlier, to the fact that countries have their own specific circumstances and market structures. I believe that the capital regulations should be harmonised further and that the aim should be to have simpler and more transparent regulations. 

An important principle is that the banks’ resilience needs to be preserved. We shall simplify but not deregulate or undermine their resilience. Historically, the experiences are clear: in a crisis, it is only stable and well-capitalised banks that continue to lend. Following the global financial crisis, the banks'’ capital has increased significantly, contributing to economic stability. The crises we have seen since then have not been amplified by the banks. On the contrary, in many cases banks have been able to help alleviate the problems by holding sufficient capital: the banks have no longer been part of the problem. If the banks become less resilient, the risk of financial crises increases, which normally leads to large welfare losses. Taking short cuts will not make the financial system resilient. It becomes resilient by our preparing for storms.

A further principle is that it is important that all actors, including the EU, adhere to international standards. Basel III has proven to have clearly positive macroeconomic effects.[10] Global financial stability is something that benefits us all. There are high contagion risks between banks and problems in one bank can quickly spread to other banks in other countries. So for financial stability in Sweden, Europe and the world, it is important that all countries fulfil the global standards. Otherwise, we risk a system of weak links, where problems in individual vulnerable banks in one country or region have serious consequences for financial stability in other countries – perhaps even globally. This is, of course, the case when crises occur at major international banks, but we know from the Icelandic financial crisis of 2008 that even serious disruptions at smaller banks can have international repercussions, for example through interconnections between banks.  

A further aspect is that the risks change over time. For example, non-banks, such as hedge funds and private credit funds, have increased their leverage over the past ten to fifteen years, creating risks that are external to the banking system but which, in stressed scenarios, may well flow back into the banking system either directly or indirectly.[11] Other examples include cyber risks, which have grown in significance, and the consequences of digitalisation. It is therefore important to be open to the possibility that regulatory frameworks may need to be adapted to these new risks. 

Some concrete proposals

The discussion on regulatory simplification focuses mainly on three areas. 

The first area is about simplifying and perhaps merging different capital buffers and making them more useful. This is known as vertical simplification [slide]. For instance, the ECB recommends merging the countercyclical buffer and the systemic risk buffer. This would make the capital more usable and thus, in line with my reasoning on the benefits of the countercyclical buffer, could increase the resilience of banks and their ability to maintain lending in more stressed situations. 

The second area is to review the many parallel requirements. This is known horizontal simplification. Let me give you two examples. 

a) Do we really need four different types of capital requirements and four (or even six for some banks) different requirements for resolution? 

b) Does Additional Tier 1 capital (AT1 instruments) really cover operating losses as intended? In this context, the ECB proposes either to phase out AT1 instruments completely or to revise the conditions.

In general, I assess that there is scope for simplification in the horizontal dimension, but this also entails difficult considerations. Do the various capital requirements and resolution requirements conflict with one another in such a way that there is a contradiction between, on the one hand, ensuring that the bank can cover losses in its day-to-day operations and still survive, and, on the other hand, ensuring that the bank can be wound up in an orderly manner once it has failed? Is it really a question of the level of the various requirements, or is it more about the process of assessing whether a bank is ‘failing or likely to fail’? I don’t have a magic formula for simplification right now, but I do think it is important that the banks have sufficient ‘going concern capital’ – in other words, banks must be able to withstand even significant losses and still continue their operations. There are potentially significant socio-economic costs associated with deciding to put a bank into resolution if it is not necessary. This could potentially undermine confidence in banks in general and create contagion risks. At the same time, it is of course essential that banks which are not viable are actually put into resolution, and that this is handled using a robust framework. Horizontal simplification of requirements calls for a very thorough analysis to ensure that these various important considerations can be weighed against one another in a balanced manner.   

As regards the Additional Tier 1 capital (AT1 instruments), I believe that there should also be room for improvement here. Historically, these instruments have struggled to cover losses in day-to-day operations, and there is currently an ongoing international debate regarding changes to them. 

The third area is about introducing simpler rules for smaller and less complex banks, known as proportionate requirements. Put simply, the EU has imposed the international standards on all banks, while other countries – the United States, Switzerland and the United Kingdom, – have much simpler but often stricter rules for smaller and less complex banks. A lot can be done to simplify the requirements and make them more efficient. It is reasonable to have more proportional requirements so that less complex banks are covered by less complex regulatory frameworks. This may mean that the requirements need to be slightly stricter, but that they will at the same time be easier to comply with. It may also be important to relax some of the reporting requirements. However, when requirements are simplified, we must not forget that some smaller banks may have operations that are complex and risky enough to warrant regulatory oversight.

In the EU, many have linked the discussion on simplifying capital adequacy rules to the process of strengthening the European Savings and Investment Union (SIU). Many SIU proponents tend to point out that simplification is a key component for the success of the SIU as it allows the banks to allocate capital more efficiently, increase lending and participate more actively in the capital markets. Simplification is a good thing and there are many good proposals within the SIU, but I am doubtful whether the proposals for simplifying capital adequacy for banks will in general lead to significantly more lending or to the objectives of the SIU being achieved in another way. Reducing the scope of national individual requirements in capital adequacy regulation could be a potential solution, however. At the same time, there are many other issues that are more important for achieving an effective capital markets union in the EU. To have an impact in SIU terms through the capital adequacy regulations, the capital adequacy requirement will probably have to be reduced substantially, but then we will also jeopardise financial stability.  

Concluding thoughts

Let me now summarise the two main themes I have raised today.

The Riksbank is looking forward to taking over responsibility for the Swedish countercyclical capital buffer and we will to a large extent continue in Finansinspektionen's footsteps, although there are several policy issues that we need to analyse further. 

Discussions on how to simplify the capital framework have started. I think that the regulatory framework definitely needs to be simplified. We can attain a simpler and better framework by increasing the releasability and usability of the buffers by abolishing or increasing the ceiling for the mandatory reciprocity and by reducing the scope for national individual requirements. This is where I think the focus should be. 

With more releasable capital, the banks can act in a way that prevents a crisis from unravelling. But it is important that the authorities responsible for the buffers also make requirements that these should be built up in good times. Such decisions are not always popular. If the buffers are to become really useful, reciprocity also needs to increase.   

Furthermore, harmonisation within the EU is desirable. This could increase the opportunities for banks to compete on a level playing field within the EU. This also increases transparency. My view is that there is scope for further simplification and harmonisation, even if we should not remove all possibilities for national adaptation. However, there are strong forces that see this as an opportunity to water down the regulatory framework to achieve better competitiveness. In all this, safeguarding the resilience of the banking system is central. We must not forget that in crises – when the banks are needed the most – only stable and well-capitalised banks continue to lend money and support creditworthy households and businesses. Weak banks do not.  

Thank you!

 

Footnotes:

[1] See, for example, Sveriges Riksbank (2017) and BCBS (2019). 
[2] See BCBS (2011).
[3] See BCBS (2010).
[4] See ESRB (2025).
[5] See Woods (2022). 
[6] See ECB (2025). 
[7] See Borio et al. (2025). 
[8] See Draghi (2024). 
[9] See European Commission (2026).
[10] See BCBS (2022). 
[11] See Federal Reserve (2025).