COVID-19: recovery and regulatory response
Conversation between Kerstin af Jochnick, Member of the Supervisory Board of the ECB, and Tim Adams, President and CEO of the Institute of International Finance (IIF), at the IIF 7th Annual European Banking Union Colloquium
17 November 2020
We have seen unprecedented action from governments and authorities – and banks – to help mitigate as much as possible the economic impact of the coronavirus (COVID-19) pandemic, but we are facing a renewed surge in cases. Meanwhile, many relief measures are running out, which will slowly sharpen the economic shock and slowdown in many cases. What can we expect now by way of impact from the second wave and what forecasts are you working with for the state of the banking sector in the EU?
There are three aspects which I would like to mention in this regard. The first is that the European banking sector entered this crisis well capitalised in aggregate terms and thereby better prepared to withstand a challenging macro environment than at the beginning of the great financial crisis.
The second aspect to highlight is that, as you rightly note, the measures taken by different stakeholders to support the banking sector during the pandemic have been extraordinary and unprecedented. The relief package that ECB Banking Supervision announced in March was designed to ensure that banks would be able to keep lending to the economy even through very harsh economic conditions, while at the same time mitigating pro-cyclicality. The measures in this package included allowing banks to temporarily operate below their required level of capital and liquidity buffers and granting some relief in the composition of Pillar 2 Requirement (P2R) capital. We also granted banks operational relief by suspending the implementation of some of our supervisory decisions for six months and extending the deadlines for the implementation of some remediation actions stemming from on-site inspections and internal model investigations. In addition, we introduced supervisory flexibility regarding the treatment of non-performing loans (NPLs) to allow banks to fully benefit from the guarantees and moratoria put in place by public authorities to tackle the current distress. Last of all, we also asked banks not to pay dividends or buy back shares until at least 1 January 2021, because we wanted to ensure that a large enough amount of capital was kept within the system so that banks would remain resilient and able to absorb losses if economic conditions deteriorated further.
These supervisory measures were complemented by other initiatives taken by various authorities to support the banking sector, including an extremely accommodative monetary stance by the ECB, the extension of payment moratoria to bank customers and loan guarantees to banks by governments in several European countries, actions taken by the European legislator to alleviate the impact from expected credit-loss provisioning under IFRS 9, and the granting of preferential treatment to non-performing exposures guaranteed by the public sector (known as the “CRR quick fix”).
Thus far, the combined impact of these measures can be assessed as positive in the sense that banks have been able to cope during the pandemic, negative feedback loops between the real economy and the financial system have been avoided, and bank lending has held up relatively well on aggregate, notwithstanding a tightening of credit standards in recent months. However, there is still a substantial degree of uncertainty surrounding the economic outlook, especially on account of the outbreak of the second wave of the virus. As recently mentioned by President Lagarde, the balance of risks to the economy is clearly tilted to the downside, particularly following the renewed containment measures taken in several European countries to counter the propagation of COVID-19.
This leads me to the third aspect I would like to highlight, which concerns bank preparedness for this uncertain outlook. The results of the vulnerability analysis that ECB Banking Supervision carried out in July suggest that non-performing loans (NPLs) could reach €1.4 trillion across euro area banks in a severe scenario that includes a second round of lockdown measures. While this is not a baseline scenario, it does give a sense of the potential magnitude of the challenge to bank asset quality which may lie ahead of us. As supervisors, we are concerned about potential cliff effects which might ultimately affect bank capitalisation when the loan guarantees and payment moratoria extended by governments to banks and their customers expire. We are thus encouraging banks to take a proactive approach to managing the expected deterioration in asset quality, without prejudice to the question of whether national moratoria measures and loan guarantees will be prolonged or not, and if so, for how long.
In facing an increasing and worsening credit risk scenario with potentially mounting NPLs, what lessons have already been learnt from the spring surge and how will the banking sector loan portfolios be affected? Are there market solutions through asset management companies or securitisation vehicles? What else can be done?
A first lesson in this regard is the one which we have just discussed, namely that banks should be proactive in guarding against potential cliff-effects in the future. A significant volume of bank loans were extended to some of the hardest hit sectors in the economy; as of August 2020, roughly 10% of all loans to the most vulnerable sectors were covered by European Banking Authority moratoria and other COVID-19 forbearance measures. Of course this does not mean that all these loans will go sour, but it does mean that banks need to make adequate arrangements as regards credit risk management. In this context, it is important that banks differentiate viable borrowers from distressed ones at an early stage, and that this effort regarding credit risk management is underpinned by a robust governance framework and backed up by prudent provisioning.
There are at least two reasons why this process of credit risk management by banks needs to be conducted in a timely and efficient manner. At the micro level, it allows the supervisor to engage with individual banks at an early stage before the negative effects on their balance sheets fully materialise. At the macro level, this aggregated process of greater visibility on banks’ balance sheets should be associated with a less protracted recessionary impact than would otherwise be the case. This is another important lesson, albeit not from this crisis but from the one prior to that, because the experience of the great financial crisis showed that well-capitalised banks that were free of balance-sheet rigidities were more likely to continue to lend throughout the cycle, and thus support a faster economic recovery, than those dragged down by high volumes of non-performing loans.
These considerations bring us to the question of what could be done to help banks dispose of their non-performing assets. In my view, market-based solutions could be a helpful addition to, but not a substitute for, adequate credit risk management by banks. We have seen that the outright sale or securitisation of NPLs have proven to be useful tools for NPL management by banks in some jurisdictions in the wake of the great financial crisis, so these remain an option going forward. In addition, a European asset management company, or a European network of national asset management companies, could also be another tool in this regard – but access to such a solution should be limited to those banks which have a viable business model, or otherwise be subject to strict conditionality. Beyond this, one must not forget that non-market solutions, such as traditional workout agreements between banks and their customers, can also be helpful for the management of bad loans.
The capital and liquidity buffers have been released, but either they are not needed yet or banks have been reluctant to use them. Could this be an area where we may see a design flaw in the framework? If so, what could be done about it and is pro-cyclicality an issue that needs to be addressed?
There are a number of important aspects to this question. The first aspect to note is that, although improvements to the system can certainly be considered, the conceptual architecture underpinning the reforms which the Basel Committee agreed upon in the wake of the great financial crisis is working largely as intended. Banks have been required to build capital and liquidity buffers, including for the purpose of helping to withstand a significant adverse shock. The partial or full use of those buffers during a sharp downturn is essential to prevent pro-cyclical behaviour by the banking sector whereby lending is excessively curtailed in an effort to preserve capital. There is also a collective action dimension to this problem, because if all banks deleveraged at the same time, the recession might be deeper and thereby have a stronger adverse effect on both the quality of bank assets and bank capitalisation than would otherwise be the case. This intuitive conclusion is confirmed by recent research undertaken by the ECB.
From a supervisory perspective, we have tried to address the issue of pro-cyclicality by encouraging banks to make use of their buffers and reassuring them that they would be given sufficient time to replenish them. This is why we have made it clear that we would allow banks to operate below the Pillar 2 Guidance (P2G) capital level and the combined buffer requirement until at least end-2022, and below the liquidity coverage ratio (LCR) until at least end-2021, without automatically triggering supervisory actions. With this “forward guidance” we have also tried to address the stigma associated with banks’ potential use of buffers.
The second aspect to note is related to the observation that, in spite of the flexibility granted by the supervisor, many banks have not used the buffers. While this has been the case, it does not necessarily follow that there are design flaws in the system – a benign interpretation would lead one to conclude that banks have not had the need to resort to such buffers yet. However, without prejudice to this, there is evidence to suggest that there are at least some banks which have indeed been reluctant to dip into the buffers – so the question is why and what, if anything, could be done about it. While it is too early to give a definitive answer to this question, I would like to highlight a couple of factors which may have been at play, namely financial market pressure and underlying bank incentives.
Concerning financial market pressure, falling capital ratios can be associated with higher funding costs or lower market valuations for banks, including as a result of the actions of rating agencies. Faced with the possibility of further ratings downgrades, some banks’ “mechanical” reaction to a worsening economic outlook could be to increase, rather than decrease, the capital targets which have been externally communicated to financial markets. If such behaviour is sufficiently common, particularly by banks in a comparatively weaker position, this may have the ultimate effect of offsetting at least partially the intended actions of the supervisor. However, the key point here is that without any regulatory relief banks could have responded in an even more pro-cyclical way, with higher capital targets associated with increased restrictions in credit supply. So I would not say that this reflects a design flaw in the framework for capital and liquidity buffers, but rather the fact that banks may face trade-offs between supervisory and market-related incentives.
Another impediment to buffer usability may have to do with the incentives inherent in the capital stacking which banks now face, because under the current legislative framework any breach of the combined buffer requirement (which sits below the Pillar 2 Guidance capital level) will be associated with restrictions on dividend distributions by banks. This means that banks may have strong incentives to keep capital levels well above this threshold, even in the presence of a broad-based recommendation by the supervisor not to distribute dividends, just to avoid an unintended breach of the combined buffer requirement and as part of their preparations for when the recommendation is lifted. In this context, we must also reflect on whether Additional Tier 1 (AT1) capital instruments are working to absorb bank losses on a ‘going concern’ basis as they were intended to do, for example when the combined buffer requirement is breached. However, it appears that banks are concerned that skipping a coupon payment on such instruments would be interpreted by markets as an early indication of solvency problems to come further down the line, akin to what would be the case in the event of non-payment of other forms of bonded bank liabilities that are only relevant on a ‘gone concern’ basis. These negative incentives on buffer usability may be exacerbated in an environment of low profitability if banks perceive that the process of replenishing any buffers which have been partly of fully used will be protracted in nature.
The fact that breaches of the combined buffer requirement are material to determine the Maximum Distributable Amount (MDA) of banks’ distributable profits has fuelled the debate in certain policy quarters of how to secure releasable buffers in a crisis. There have been a number of different proposals put forward to this end, but the basic idea common to most of them is that a re-balancing of the structural versus cyclical elements of banks’ capital stacking would give macroprudential authorities greater capacity to act in a countercyclical manner. At the same time, this buffer release would also reduce banks’ incentives to act pro-cyclically because the MDA trigger would be commensurately lowered.
While I welcome such a debate, I think it is premature to draw any firm conclusions on this question at present. We should wait for the crisis caused by the pandemic to be over before discussing the extent to which tools afforded to micro and macroprudential authorities in relation to banks’ capital stacking should be recalibrated. Moreover, I also think that the current crisis is giving us an incomplete view of the degree to which the entire Basel III capital framework is effectively working as intended, because some of the elements have been insufficiently implemented, if at all. For example, if cyclical macroprudential buffers had been built up more during the macroeconomic upturn in the run-up to the pandemic, the capacity for macroprudential authorities to act counter cyclically in a downturn would have been greater.
Whilst the strength and depth of capital and liquidity stand at high levels among European banks, the weak point remains low profitability resulting from structural factors. Achieving scale and cost synergies through consolidation would be one way to improve. Can we expect a wave of consolidation soon – domestically and/or cross-border? If not, what potential remedies do you see?
Low profitability is indeed a significant vulnerability of our banking system. Its causes are diverse in nature, with high cost structures, still significant legacy (non-performing) assets in some cases, weak income-generating abilities in others, and excess capacity in some banking segments all weighing on bank performance. Coupled with subdued macroeconomic prospects, these factors push down bank valuations and lead to a situation where most banks are not earning their cost of capital at present. Lacklustre profitability can also condition banks’ response to the digital challenge, because investment in digitalisation is a significant sunk cost to be paid upfront by banks whose benefits, even if substantial, may only be reaped over time. Some bank laggards on the digitalisation front that postpone their investments either out of choice or obligation can then see a further decline in their subsequent performance.
As supervisors, we had been conveying the need for banks to adapt their business models to the new realities of the banking business (for example, related to the implications of the low interest rate environment) well before the outbreak of the pandemic. The need to adapt still exists, but it is fair to say that this has been hastened by the COVID-19 shock because many of the outstanding challenges which have a bearing on low profitability, including those related to digitalisation, have been brought to the fore during the current crisis.
We thus see consolidation as one way to help banks deal with the profitability challenge through scale and cost synergies, while also addressing the excess capacity which still exists in parts of the banking system. Judging by the recent increase in consolidation involving large and mid-sized players in some countries, one could arrive at the conclusion that this view is shared in at least part of the banking community. However, we should clarify in this regard that our role as supervisors is neither to push for consolidation nor to stand in its way. In other words, our task is to ensure that new entities resulting from business combinations have sustainable business models, comply with prudential requirements, and have sound governance and risk management arrangements in place that ensure an adequate coverage of their risks. In order to make our supervisory actions more predictable and help bring about prudentially sustainable consolidation projects, we published a draft Guide on our supervisory approach to consolidation in July this year, clarifying our expectations in this domain. We are now processing all the comments received and will publish a final version of the Guide soon.
At the same time, however, we should be aware that there are still a number of elements standing in the way of an integrated banking market across Europe, and that until these are addressed it is unlikely that cross-border consolidation among banks will be given significant impetus. The true test of banking union is therefore yet to come, because the European banking market remains fragmented along national lines. It was – and to a certain degree, still is – widely believed in host countries that, in times of crisis, parent banks will protect their own interests and home authorities will prioritise their own national objectives. The long-standing reluctance by host authorities to allow cross-border banking groups to manage capital and liquidity at the group-wide level increased during the great financial crisis, with ring-fencing measures appearing to be the favoured crisis management tool to limit the fallout in a purely domestic context. Although much progress has since been made to break down suspicions between host and home authorities, including through the establishment of a Single Supervisory Mechanism with the ECB at its core, part of this ring-fencing attitude persists, and has thereby conditioned progress towards the establishment of banking union.
In order to overcome some of these obstacles, and pending more far-reaching reforms such as the establishment of a common insurance scheme for bank deposits across the banking union, my colleagues on the ECB Supervisory Board, Andrea Enria and Edouard Fernandez-Bollo, have recently put forward a number of innovative proposals to make progress towards a more integrated banking market, including by introducing adequate incentives and safeguards to enter into group support agreements and linking those group support agreements to recovery plans.
How do you see the chances of completing banking union during this crisis? Are we any closer to that or have the massive support packages coming from governments and the EU made the need for sharing risk less imperative and counterintuitively strengthened the sovereign-bank nexus?
This is the sort of question that divides people depending on whether they see the glass as half full or half empty. As you might expect, I count myself among the first camp, but I am conscious that the concerns of the other group need to be taken seriously. Let me briefly elaborate on this thought.
In my view, banking union has withstood the test brought about by COVID-19 very well. From an operational point of view, this was the first time that we had a single banking supervisor rolling out a relief package for banks across the euro area. This allowed for measures that were unprecedented, both in scale and effectiveness, to contain the immediate economic fallout arising from the pandemic as far as possible. Part of the reason for this “success” is that the home-host coordination problems which were prevalent prior to the establishment of the Single Supervisory Mechanism have now been fully internalised.
From a conceptual point of view, the crisis has also shown us the benefits of having pan-European structures to regulate and oversee banking activity in its different forms. By applying the stricter and more harmonised regulation that was put in place after the global financial crisis, and by developing practices and norms to bring the entire system to a higher common supervisory standard, ECB Banking Supervision has made sure that the European banking system as a whole is better poised to withstand severe crises such as this one.
So there is the “glass half full” view; let me turn to the “glass half empty” one. A first word of caution in this regard – there should be no room for complacency. The fact that banking union has successfully weathered the challenges brought about by the COVID-19 crisis thus far does not necessarily imply that it will continue to do so indefinitely – whether at other stages of the crisis or in similar situations. It’s essential to continue building the pan-European architecture envisaged by political leaders when establishing banking union. Beyond the structural impediments that still need to be removed – some of which I mentioned earlier – we ultimately need to make banking union fully operational by putting in place a European deposit insurance scheme.
A second word of caution concerns the strengthening of the bank-sovereign nexus, as you have correctly noted in your question. Though appropriate for dealing with the current crisis situation, extending government guarantees to banks and decreeing payment moratoria for their customers has indeed reinforced the linkages between domestic banking systems and their respective sovereigns – though hopefully only in a temporary manner. Going forward, we will need to guard against the risk that the very same issue which banking union was designed to tackle in the first place does not actually come out of this crisis reinforced. This calls for a close monitoring of the exit out of the payment moratoria and loan guarantees which have been extended to banks across Europe, as well as a focus on banks’ exposures to local or central governments in a context in which public debt issuances have been rising.
Covid-19 lockdowns have led to a very extended period of working from home; banks and staff have adjusted to this new paradigm with remarkable agility, but how have the supervisors adjusted and how has the suspension of on-site inspections changed the way the supervisory teams work?
Just like the banks it supervises, ECB Banking Supervision has had to adjust both its operations and its supervisory approach to the new reality of social distancing and travel restrictions. For example, the ECB Supervisory Board has held all its meetings via telephone or video conference since mid-March this year, and the bulk of our supervisors have been working remotely since that time, too. Overall, I am impressed that we were all able to adapt so efficiently to the remote working mode in the middle of a crisis.
Insofar as on-site inspections are concerned, we initially suspended the implementation of some of our supervisory decisions for six months and also extended the deadlines for the implementation of some remediation actions stemming from on-site inspections and internal model investigations. These measures, which were part of our initiatives to grant operational relief to banks, were announced in March. Thereafter, we moved into a set-up in which on-site presence was substituted by off-site scrutiny, implying that on-site inspections/investigations were converted into off-site inspections/investigations to the extent possible. Inspections requested by our joint supervisory teams are carried out remotely by supervisors from both the ECB and the national competent authorities (NCAs), working collaboratively and in mixed teams just as we did before the outbreak of the pandemic. Our off-site approach has also been complemented by the updates which we have regularly requested from banks on risks that may be particularly heightened during this crisis, so that pressing vulnerabilities are flagged early on and we are able to address them promptly. My overall impression is that both the independence and the quality of our supervisory scrutiny have not been compromised by these changes.
In the future, once travel restrictions are fully lifted and it is safe to travel again, we will be drawing lessons from this new off-site approach to inspections, and will probably complement the traditional on-site inspection toolbox with some remote techniques that are currently proving very effective.
This pandemic has significantly accelerated the digitalisation of the industry. What issues have emerged from a supervisory perspective in this context?
Indeed, the pandemic has been associated with a decisive growth in the demand for digital products by bank customers. In my view, this feature is likely to remain part of the “new banking normal” once the pandemic is over, meaning that banks will need to have competitive digital products on offer if they are to survive. This will require banks to further revise their cost structures and channel their investments towards becoming digital.
With this objective in mind, we have been engaging with banks by carrying out pulse checks on both their transformation processes and their take-up of innovative technological solutions. This involves looking at their innovation strategies, IT budgets and overall allocation of human and financial resources to digitalisation.
The transition to the “new banking normal” in the digital realm will also imply that areas such as cyber risks, IT risks and operational resilience will demand greater supervisory attention from us than has previously been the case. So far banks’ IT infrastructures have stood up well to these challenges. Almost all banks under our direct supervision have managed to continue providing services to their customers without major disruptions or setbacks to their operational resilience. Going forward, banks should be ready to do away with legacy IT systems and to streamline internal platforms that are resilient enough to withstand heavy reliance on remote working and provision of services. Overall, the ever more digital environment we are likely to experience in the period ahead will bring both new opportunities and new challenges for regulators and banks alike, as they work to ensure that the banking sector remains safe and sound.
Can you comment on the recommendation banning dividend distributions and share buy-backs? When and under what conditions could this policy be reversed?
The overriding objective of this measure was to maintain precious capital resources in the system so that banks could absorb losses arising from the pandemic while also being able to continue lending to the real economy. We decided on the first recommendation back in March, initially for a six-month period, and renewed it for another three months in July, so that restrictions on dividend distributions and share buy-backs are now recommended until 1 January 2021. We will be revising our recommendation in December, so no decision has been taken yet. While I cannot anticipate the decision that we will take, I can share with you two key elements which have influenced our thinking on this matter until now.
A first element has been the degree of uncertainty prevailing in the macroeconomic outlook. The initial recommendation was taken following the outbreak of the pandemic in Europe and at a time of high uncertainty surrounding both the consequences and the duration of the COVID-19 shock.
A second and related element has been the extent to which we had visibility over banks’ balance sheets, particularly as regards the trajectory of asset quality – a key indicator used to inform realistic and reliable capital planning by banks. Our decision to extend the recommendation in July erred partly on the side of caution amid very diverse estimates by banks in this regard; this variance also extended to other areas such as the cost of risk and loan classification.
We will adopt this stance until we are satisfied that the uncertainty no longer prevents banks from producing reliable estimates of how their balance sheets will be affected by the pandemic. We have been closely monitoring the economic environment, the stability of the financial system and the reliability of banks’ capital planning. Taking all these into account, we will decide on whether or not we would like to see banks continue to refrain from paying dividends or buying back shares beyond 2020.
Let me add that, while I imagine that our recommendation to restrict dividend distributions has not been popular with many people in the audience today, I am also confident that it was the right decision to take, both in a prudential sense as well as from the point of view of corporate social responsibility. The fact that banks are seen as part of the solution to this crisis (as is the case at present), rather than as part of the problem which led to a crisis situation (as was the case during the great financial crisis), is a positive development which should go some way to repairing the general public’s perception of both bankers and the banking business.
Could you comment on the role of the ECB as a banking supervisor with regard to climate-related financial risks? What are the main objectives of banking supervision in this area? Do you envision helping the banking industry accelerate its transition to greener asset allocation?
In a basic sense, banking supervision entails monitoring banks’ operation and performance to ensure that these are conducted in a safe and sound manner and with due regard for the prevailing rules and regulations. It is clear that climate change will be a point of growing supervisory attention going forward, both from the perspective of financial risk as well as from the point of view of compliance with applicable legislation.
The first task for banks and supervisors alike is thus to assess the prudential risk dimension of climate change. So far, banks seem to have engaged with this topic mostly from a corporate responsibility perspective. They have devised sustainability strategies that merely outline how they can positively contribute to climate and sustainability objectives, but often lack a more applied angle. Going forward, banks need to complement this engagement with a more traditional risk-management approach to effectively understand how they can be financially exposed to and impacted by climate risks, so that these are incorporated into their risk-assessment frameworks.
To assist banks in this task, ECB Banking Supervision has published a “Guide on climate-related and environmental risks”, intended to raise banks’ awareness and to better prepare them for the future. The Guide, which was released in May and was subject to a public consultation that ran until end-September, also provides clarity on the ECB’s understanding of the safe and prudent management of climate-related and environmental risks under current prudential rules. We expect to publish the final version of the Guide very soon.
Looking further ahead, we plan to enter into a supervisory dialogue with banks to foster a common understanding of the importance of adequately managing and disclosing climate-related risks in 2021. While we do recognise that the measurement of climate-related and environmental risks is still evolving, we expect banks to start building their capacity and adapting their practices as of now.