Transcript of the media briefing on July 28, 2020 (with Q&A)
Andrea Enria, Chair of the Supervisory Board of the ECB,
Jump to the transcript of the questions and answers
Frankfurt am Main, 28 July 2020
Andrea Enria, Chair of the Supervisory Board of the ECB
Good morning and thank you for joining this press briefing in the middle of the summer. Today we published our first assessment of the potential vulnerabilities of the euro area banking sector in the face of the coronavirus (COVID-19) outbreak, together with a comprehensive set of communications on distributions, remuneration, capital and liquidity buffers, and ongoing supervisory activities. We hope that today’s publications, taken together, will help to provide as much as clarity as possible under the current circumstances of extreme uncertainty, allowing banks to better plan for the foreseeable future. Let me move quickly to the results of our assessment and the different types of measures announced.
Our vulnerability analysis is based on two macroeconomic projections published by the ECB in June 2020. The ECB’s macroeconomic baseline projection, which is the ECB’s forecast for 2019-22, is our central scenario, and the ECB’s alternative severe projection is our severe scenario. In the central scenario, we have GDP falling by 8.7% this year, and an almost complete recovery by 2022, so the cumulative GDP development over the three years would be ‑0.8%. Under this scenario, the aggregate Common Equity Tier 1 (CET1) ratio of the euro area banking sector would fall by 1.9 percentage points, so from 14.5% at the end of 2019 to 12.6% in 2022. Our opinion is that, under this scenario, the euro area banking sector would remain well capitalised overall and could continue to fulfil its core function of lending to the real economy. However, the macroeconomic outlook remains uncertain. In particular, according to ECB economists, a severe scenario, whereby GDP would fall by 12.6% this year and there would be a much slower, flatter recovery over the next two years, remains plausible. The cumulative impact would be -6.3%, which would take us to 10.3% below the pre-COVID-19 baseline scenario – so, quite a severe scenario. In this case, the aggregate CET1 ratio of the euro area banking sector would decline, on average, by 5.7 percentage points, reaching 8.8% by the end of 2022.
Of course, such a pronounced reduction in the sector’s own funds would prove challenging, and several banks would need to take action to continue meeting the minimum capital requirements. Still, the overall capital shortfall of the sector would be contained. We believe that there are two elements that stand out about the current environment. The first is that the macroeconomic outlook remains very uncertain, which makes it difficult to foresee the impact of the current crisis on banks’ asset quality. What is most relevant for us is that both we and the banks still find it difficult to accurately estimate the capital trajectory, especially after the moratoria come to an end. In this context, we think it is our duty to err on the side of prudence and take measures to mitigate the risk of deleveraging, the risk of procyclical behaviour, on the part of the banking sector.
This is the background to our decision to extend our recommendation to banks to refrain from dividend distributions and share buy-backs until the end of 2020. We have also sought to clarify to banks that they will have enough time to replenish the buffers, and to provide a clear indication of how this would be done. This is important because, so far, banks have claimed that they find it difficult to use the buffers because they have no clarity as to what the ECB would require when it comes to the timeline for replenishing their buffers. So our decision on distribution has been guided by the need to preserve capital in the sector, as a way to ensure sufficient loss-absorbing and lending capacity. I’m aware that this won’t be particularly well received by investors, but let me repeat that this recommendation is exceptional and temporary in nature. We will review the decision in December, and if we can estimate the impact of the macroeconomic outlook on asset quality with more certainty, and if our supervisors are convinced that banks’ capital projections are sufficiently reliable, we will repeal our recommendation and move back to our ordinary assessment of planned distributions on a bank-by-bank basis. But we prefer to be prudent today rather than having regrets tomorrow should the overall conditions of the economy deteriorate further.
Another point that we clarified today is that our recommendation refers only to cash dividends, so banks can pay scrip dividends, i.e. dividends in shares, because they don’t affect the capital position of the banks. Our decision to extend our recommendation on distributions is also consistent with the recommendation of the European Systemic Risk Board (ESRB). As the recommendation of the ESRB also refers to bankers’ remuneration, today we also sent a letter to banks under our direct supervision, recommending that they take an extremely moderate approach to variable remuneration payments until the end of 2020.
Buffer usability has been a pillar of our crisis response. ECB simulations that we published today show that, if banks decided to use their capital buffers, there would be clear benefits for the economy and for the banks themselves, without any noticeable negative effect on the banks’ capital position. Using the buffers would support higher GDP growth and, as a result, credit losses would be lower and bank profits would be higher.
We have clarified today that banks will not need to start replenishing the buffers before the peak of capital depletion is reached. In any case, banks will not need to comply with the Pillar 2 guidance and the combined buffer requirement any sooner than the end of 2022. We hope that the overall situation improves and that in 2021 we will be able to conduct the regular stress test under the aegis of the European Banking Authority (EBA). This will enable us to set up new Pillar 2 guidance at the end of 2021, and then define specific timelines, on a bank-by-bank basis, for reaching the capital levels set out in this Pillar 2 guidance. Our supervisors will work with the banks to decide on the most suitable path to achieve these levels – if they use the buffers, of course. In particular, our supervisors will take into account the profitability of each bank and how far each bank is from the buffer target. This means that we will rely very much on internal capital generation. The same will apply for the combined buffer requirement, so there would be the same timeline and process for banks that breach the combined buffer requirement.
A similar approach will be also applied for the liquidity buffer, i.e. the high-quality liquid assets which form part of the liquidity coverage ratio (LCR) requirement. Here, banks will not be required to restore compliance any earlier than the end of 2021. In any case, specific adjustment paths will be designed on a bank-by-bank basis, taking into account developments in the conditions in funding markets. We hope that this provides sufficient clarity to banks that we are not going to switch the buffers on again and put them into a difficult situation if they use them today.
Finally, let me mention the operational relief measures that we started in March. Today we have also provided an update on those measures. For instance, we announced that banks will not be required to submit until March 2021 their plans for reducing non-performing loans. This is necessary because it is not sufficiently clear how far asset quality will deteriorate in the months to come. On the other hand, we also recommended that banks with high levels of non-performing loans continue to focus on reducing them to sustainable levels and take all available opportunities to do so. We’re also sending a letter to all banks to make sure that they step up their operational capacity to deal with the expected increase in impaired assets. It is a so-called “Dear CEO” letter. The six-month suspension of pending supervisory decisions has not been renewed. This means that from October 2020 onwards banks will be expected to resume taking remedial actions based on findings in SREP assessments, on-site inspections and internal model investigations. Supervisory processes for adopting new decisions will also restart. For instance, we suspended some decisions related to the targeted review of internal models, and these will start again from October 2020.
The sweeping reforms that we made in the last decade have been effective in strengthening the banking sector, and have made the banking sector sufficiently resilient to withstand a shock of unprecedented magnitude. There is, however, still some degree of uncertainty regarding future developments and the effect that these will have on banks’ balance sheets. Our supervisory measures and actions are and will continue to be focused on support for the banking sector and on the support that the banking sector needs to provide to the real economy. I look forward to answering your questions. Thank you.
* * *
In your press release you clarify that, considering the current circumstances and in order to avoid subjecting the banks to an additional operational burden, for your analysis you used data that were already available. I was wondering if you have any specific recommendations for the Greek banks, given that deferred tax credits (DTCs) make up a disproportionately large portion of the capital held by the Greek lenders. The fact that no regular tax audit has been carried out on the Greek systemic banks since 2010 could, during this period or in the forthcoming period, lead to even bigger provisions and additional capital needs.
I appreciate your question, but this is not really a question for today. Today the focus is not on specific banks or banks in specific countries, but more on the overall euro area banking sector. The issue of DTCs in Greek banks is, of course, something which is duly considered by our supervisory teams. They continuously engage with each of our banks and closely monitor their capital positions. I know that there are initiatives under discussion in Greece, but at the moment capital projections are very challenging for everybody because of the uncertainty we are in.
First, you say in your presentation that, in the severe scenario, several banks would need to take action to maintain compliance with minimal capital requirements. What kind of action would they need to take, and is it credible that they’d all be able to take it, given that you’re talking about a huge capital decline of 7.6 percentage points for diversified banks and so on, so they’d be in a very weak position?
The second question is: your previous dividend ban was criticised because it left a loophole for banks’ subsidiaries. For example, Bankia, which has a parent that is also a bank, was able to pay a dividend. 40% of Bankia shareholders received a dividend because it wasn’t caught by your recommendation. Why does that loophole exist, and have you managed to close it?
The point is that as soon as we see the situation of the banks deteriorating and their capital position becoming weaker, we would of course expect them to take action to, first of all, deal with their asset quality deterioration. And that's why we also issued a letter today to banks asking them to start preparing themselves to actively manage their distressed asset portfolios, as these will likely increase. Second, we would expect banks to take action aimed at strengthening their capital position. We have recovery plans that banks develop, in which there are a number of triggers that would prompt them to take action to reduce their asset size or beef up their capital position. Eventually, they could also consider tapping markets for certain capital instruments. I notice that markets for Additional Tier 1 and Tier 2 are particularly open, with good conditions, so banks could also take this avenue. In general, we’d expect banks to take any action that can improve their capital position. This will be discussed with individual banks on the basis of their recovery plans, of course.
On your second question, I would not agree with your characterisation of the issue as a loophole. There is no loophole. It was a deliberate decision of the Supervisory Board to apply our recommendation only at the higher level of consolidation. If there is capital which is distributed within the group, the capital remains in the group, and remains available to absorb losses and support lending. So there is no loss of capital, no exit of capital, from the sector. The internal distribution of capital within banking groups in the euro area is not the core focus of our attention. Of course, if there is a specific fragility in a local subsidiary that might require an idiosyncratic intervention, this is something that we would consider on a case-by-case basis.
Two questions from my side. The first one: is Pillar 2 guidance dead? The reason I ask is if you are allowing banks to pay dividends, even if they're below their Pillar 2 guidance, what's the point of having Pillar 2 guidance?
Then the second question is a technical one. You've said before that if banks don't comply, then you do have legal follow-up actions you can pursue. Can you explain those follow-up actions so we understand them better? What happens if a whole bunch of banks don't comply? It seems like a lot of work to go bank-by-bank and make sure they fall in line.
Well, the short answer to your first question is no. The whole point of our publication today is to say how the Pillar 2 guidance will be set in place again. The point is how do you construct the path for rebuilding the Pillar 2 guidance? The key message is that banks will not be rushed into doing it in a very short period of time, and also that if they have a clear path for getting there within a certain time frame, which has been agreed with the supervisors, they will also be allowed to pay some dividends compatible with this adjustment path. So Pillar 2 guidance is far from dead. Actually, it is a sort of textbook example of how buffers need to be manoeuvred during a crisis: enabling banks to use them and then creating a clear path for repletion.
On your second question, yes, our recommendation on dividends is exactly that, a recommendation. So we expect the banks to comply with it, but it is not in itself a legally-binding requirement. However, we do have the possibility to issue binding requirements related to distributions on a bank-by-bank basis, and we would not hesitate to do that if banks do not comply with our recommendations. I am very pleased that banks have followed our recommendation issued in March, and I'm confident that they will do the same now as well. We are talking about a recommendation that applies until the end of the year. It is a highly uncertain environment. I think it is in the interest of the banks to wait until there is more clarity on the future path in terms of capital and asset quality before taking firm decisions on distribution.
I have a question with regard to the dividends. So why have you chosen this indiscriminate approach to dividends? There are cases where the dividend would amount to a very low sum compared with already-reached capital levels, so it doesn't really make a big difference to the lending capacity. So I have the feeling it's more of a symbol than being really necessary for all.
The second question is, is it a little bit of a lesson from the last financial crisis, where the United States forced the big banks to take money from the government, so that also the weak are protected on the shoulders of the strong?
It's clear that a bank which has very significant capital buffers above minimum requirements would be particularly annoyed about being prevented from benefiting from its strength and paying dividends to its shareholders. Also for us as supervisors, it's more natural to go back to a situation in which you differentiate across banks, and you try to preserve capital at weak banks, but enable banks with sufficient capital buffers to distribute dividends and be more profitable and attractive to investors. However, we are now in a very peculiar and different situation. Suppose that we set a specific capital threshold that you need to have to be able to distribute dividends. All banks would try to remain above that capital threshold. They would probably deleverage in order to avoid going below that capital threshold, and this could have an impact on their willingness to lend to the real economy, for instance.
Also, there is such radical uncertainty that it's very difficult to distinguish between banks. For instance, one of the drivers of the risks for banks is the concentration of exposures to sectors which are particularly affected by the pandemic. So you can have a bank which appears to have a very healthy capital position, but might be significantly exposed to a sector which is particularly exposed to the COVID-19 outbreak. The capital position of this bank could deplete very fast. It would also be very damaging for us if these banks were allowed to pay dividends and after a while found themselves in breach of requirements. So due to the uncertainty and the need to prevent any incentive for banks to deleverage and maintain buffers in place, I think it is appropriate at this stage to have the one-size-fits-all recommendation. As I mentioned before, it's an exceptional and temporary measure, and we are very willing to go back to the bank-by-bank approach as soon as this is possible.
The first question is about the additional action that you refer to when you say that, if the economy develops along the lines of a severe scenario, authorities must be ready to take additional action to prevent, effectively, a credit crunch. So what are you referring to?
The second question is about how the two decisions sit together. Isn't the decision to ask banks not to pay dividends redundant if you also say that you are not going to look at the capital conservation buffer?
On the additional actions, we don't have any specific recommendation in mind at the moment. We are just noting that, if the situation starts deteriorating and the capital depletion is quite material, banks would in all likelihood start deleveraging. Governments have taken actions to support borrowers, to support households and corporates. They have provided guarantees so that borrowers can continue to have access to lending, and moratoria for banks' customers. So one possible line of action — also mentioned in the results of the vulnerability analysis – is that governments might consider prolonging or extending the scope of these measures for borrowers, which would, of course, also protect the banks from the impact of the crisis. In the results of the vulnerability analysis, we also mention the positive effect of the liquidity measures of the targeted longer-term refinancing operations (TLTRO), so there are also central banking measures which have been particularly strong in supporting the banking sector. Eventually – and we already have this in place in some countries – authorities could consider other types of support measures, such as asset relief schemes, or capitalisation if need be. It's not the case right now, but these types of measures are there and there is a regulatory framework in place for dealing them.
On the question on the measures’ redundancy: no, I think that the two measures are not contradictory. Actually, they sit very well together. At this point the message to banks is they should keep capital in the system to absorb losses and to lend to the economy, lend to households, small businesses and corporates. That is the focus that we need to have at present: preventing capital from exiting the system and focusing this fire power on necessary loss absorption and lending capacity.
Korbinian Ibel (Director General Microprudential Supervision IV): You'll find the information that the Chair was referring to in slides 21 and 22 [in the results of the vulnerability analysis]. You will see what would happen if the deadline for the support measures was extended by six months. This would provide banks with relief amounting to €18 billion. You'll see on slide 22 the effect of public guarantee schemes and of lower cost of funding due to the TLTRO III, and there you'll also see that there is additional power in these measures.
I actually have a question on the state guarantees. In that slide it mentions extending them by six months. So do you foresee a problem at year-end if that doesn't happen?
Second, will you call on governments to extend them if the economy does worsen?
Korbinian Ibel: Slide number 21 in the results of the vulnerability analysis is not meant as a policy suggestion, but is rather meant to show that the analysis we did is, of course, sensitive to different assumptions. If the economy worsens, then maybe there's an expectation that the guarantee schemes would be extended. We wanted to show here that this would indeed have a positive effect according to our analysis.
I just have a couple of quick questions. When you say that the dividend recommendation will be reviewed in the fourth quarter, can you explain what that means in practice?
Also, have you considered how the dividend decision is going to impact banks' share prices?
Finally, once banks start thinking about replenishing the capital buffers do you expect some of them to consider mergers & acquisitions (M&A) as a way to replenish such buffers?
Again, the recommendation has been extended until 1 January, which means that in the fourth quarter we will review it. One key point will be to understand whether we can estimate the impact of the macroeconomic environment on asset quality at that stage. In addition, [we will have to see] how reliable our supervisors think the banks’ capital plans are. If you have clear visibility on banks’ capital trajectory, on a bank-by-bank basis, then you can move to the business-as-usual of discussing with the banks their dividend payments on the basis of the capital trajectory. But if you still don’t have visibility, as is the case now, then it is very difficult to differentiate across banks. So that’s what the review will focus on.
On the impact on banks’ share prices, we have seen that there has been an impact on market prices, as expected. We know that investors have not been particularly pleased by our decision. But again, I think that this is a necessary action to take at this time of heightened uncertainty, where it is important to ask banks to focus their capital resources on lending and on loss absorption.
Finally, on M&A and the path to review the buffers, I’m not sure whether there is a close connection between the two. In general, we think that consolidation could be an important element of a strategy for the European banking sector to recover profitability and eliminate excess capacity. Indeed, that could be something that might be explored in the recovery phase of the current recession.
Do you see in this analysis any specific need for consolidation, not a general recommendation to consolidate, but a specific need?
It’s not this [vulnerability] analysis that highlights the need for consolidation. The need for consolidation stems more from the structural weakness of the banking sector in the euro area, where cost-to-income ratios are very high and banks have relatively little ability to invest in new technologies and restore profitability to sustainable levels. So we think that, to some extent, this has a bearing on the situation of banks, and their weaknesses vis-à-vis the vulnerability analysis. For instance, one of the results highlighted in our findings is that more profitable banks are better able to take on the shock of the severe scenario. This is not a surprise, of course. The generation of profits is the first line of defence for a bank in difficult macroeconomic conditions. So indeed, consolidation is a useful tool to restore profitability, to deal with cost inefficiencies that we still see in the sector.
We are also seeing that banks have been quite efficient in moving to a new operational mode, working mainly remotely, with many branches having closed during the lockdown period. This has shown everybody, both banks and their customers, how far you can go in terms of relying on new technologies to provide banking services. In a sense, this provides food for thought for bank managers to understand how they could reshape their provision of services going forward. Maybe consolidation could fit into that picture, because it could provide an opportunity to cut end-of-life IT systems, re-establish new and more fit-for-purpose IT frameworks, develop technologies and cut branches where distribution networks overlap. In particular, it could be an opportunity to restore cost efficiency and profitability, which is an important safeguard in a situation of deep recession.
Korbinian Ibel: If you go to slide number 14 [in the results of the vulnerability analysis], then you see waterfall charts that go from the starting point to the final CET1 ratio for the three different scenarios we calculated. You see that administrative expenses have a very high negative impact and that, of course, net interest income and commission income have a very positive one. These elements basically comprise the cost-to-income ratio of a bank. So if a bank has a high cost-to-income ratio and is unprofitable going into an exercise like that, it has no buffer to compensate for the increased credit risk and market risk. If it has good profitability then it has a buffer to compensate for that, and this buffer then has to withstand the crisis and the stress we apply to it. So in this way you can see on slide 14 exactly what the Chair just explained.
I have a question regarding the implementation of Basel III in Europe. The EBA proposed a relatively tough implementation. Do you support this or are you afraid of disadvantages compared with the United States?
We are strongly supportive of the stance of the EBA, and we have ourselves supported the faithful and timely implementation of the Basel package. Of course, timely now also means in line with the delay that has been envisaged by the European Commission. But we think that, all in all, the package strikes a good balance. I think that most of the measures which have been introduced with the new international standards come from analyses which have been conducted in Europe by the EBA and the ECB. They are all aimed at closing loopholes and fixing problems which have generated level playing field issues across banks. One issue which has been hotly debated is the output floor, which would constrain the extent to which banks could benefit by relying on internal models as opposed to the standardised approaches. This has been a very controversial issue, but I think that this requirement will be a part of the overall package and final solution – including our clarification last year as to how it should be implemented in the European Union.
A further question on dividends. You say in your recommendation that you will further evaluate the economic situation, and consider a suspension of dividends beyond 1 January 2021. That could mean several years without dividends, potentially. Do you think it’s likely that this recommendation will be further extended? When would you take that decision, and on what basis?
I didn’t say that there is any expectation of an extension. We will review the decision in December, ahead of the deadline and well ahead of the season for the payment of the 2020 dividends. The key points for us are visibility on how the macroeconomic situation will affect the asset quality of banks, and the reliability of capital projections. These are the two main elements, which actually are two sides of the same coin, to some extent. As long as our supervisors are sufficiently confident that the capital projections the banks put forward are reliable, they will be able to differentiate clearly between banks which have strong capital buffers and can give sufficient reassurance that they will be respecting the [capital] requirements in the medium term while paying dividends, and other banks which are maybe not in such a position. As long as this is possible, we will repeal the recommendation and go back to the ordinary world in which banks are free to pay dividends.
Let me also say that, at the beginning of this year, we significantly increased the transparency of our Pillar 2 process. If I had to express a preference, I would like to move to a situation in which you have full disclosure of the requirements and buffers which are required of each and every bank. Then you know that there is a certain management buffer that banks are expected to keep on top of that, and then banks would be free to adopt a dividend policy or conduct buy-backs as they please, if they are above that threshold. In normal times that would be my preference. As I mentioned before, these are not normal times, and the decision that we have taken now, which applies to all banks across the board, is of course a very exceptional and temporary measure. So I hope we will go back to normal pretty soon, not only for the banks, but also for our daily lives. As soon as this happens we will be ready to review the recommendation.
How would you characterise or describe the responsibility or role of the ECB in the Wirecard case? In what way was the ECB involved in the decision not to consider Wirecard as a financial holding, and what consequences, if any, will the ECB Banking Supervision draw from that scandal?
I wouldn’t enter into great detail. As you know, Wirecard is a less significant institution. We do not supervise it directly. Of course, there are moments in which we engage with our colleagues in the national authorities, in this case the BaFin, for specific procedures. This was the case for a qualifying holding procedure initiated by Wirecard within the framework of the restructuring of the group. So that’s the involvement that we have had with the case. We have worked jointly with BaFin on this topic, but it is not our responsibility to define an entity as a financial holding company. We reviewed the [restructuring] case as a part of the qualified holding procedure.
In terms of lessons, I think it’s definitely an important case. I would also say that last year, within our report on less significant institutions, we published an interesting part which focused on those new fintech start-ups which specifically focus on the provision of services based on new technologies, and there are specific issues which emerge there. Of course, we focus very much on the banks, but I think that there might also be lessons for the technology companies which provide financial services outside the remit of a banking licence.
My first question is in relation to the uneven impact that the crisis is expected to have across jurisdictions in the eurozone. I was wondering whether you’ve identified this uneven impact in banks across different countries?
My second question would be whether you foresee that measures such as the recovery fund approved last week – this coordinated fiscal response and the protection you could provide to weaker economies in the euro area – could be part of your analysis and could affect your analysis in the coming months and years.
Well, the analysis we conducted is not focused on banks in specific countries or on a specific cluster of banks. We gave indications of the impact of the different scenarios on banks with different business models. But what I would say is that this is a peculiar crisis where the impact depends on a number of factors which are not necessarily linked to geographic distributions. Of course, banks which are starting with a higher level of non-performing loans will be in a particularly challenging position, but this is not the only element in the picture.
In terms of the recovery fund, I think that this is an important piece of good news for the European Union and for the euro area, insofar as it will provide a very strong and joined-up fiscal response to the crisis. To some extent, we have already factored into the scenarios the monetary policy and fiscal position that was known at the time. I expect that the more support you have for the economy, the better the economic outlook will become, and therefore there will be less concern for the situation of the banks and their balance sheets.
Korbinian Ibel: I would like to confirm exactly what you said. The recovery fund would most probably have a big impact on a future update of the scenarios, which would then translate into an effect on the banks. So this would be the transmission mechanism. I guess that this will be incorporated into the next ECB staff projections and then it might change the scenarios. If you ran the exercise again on an updated scenario you would, of course, get different results. Among the jurisdictions, I can confirm that we have the data, but the distinction between business models, which you see on slide 17, is much more significant than the distinction between countries. This is why we did not include the latter in the presentation.