Banks after the end of “low for long”

Speech by Kerstin af Jochnick, Member of the Supervisory Board of the ECB, at the 28th Annual Financials CEO Conference, hosted by Bank of America

London, 20 September 2023

Introduction

It is a great pleasure for me to be here with you today, and I would like to thank the organisers for their kind invitation.^[1] When thinking about topics that would keep an audience of prominent bankers sufficiently engaged, a quote from Hartley Withers, the English financial journalist who was editor of “The Economist” during the early part of the 20th century, came to mind. He once said that good banking is produced not by good laws, but by good bankers. While this characterisation of the banking industry was probably accurate back then, I hope you agree that, nowadays, good laws and efficient supervision are also essential for resilient and prosperous banks. With this in mind, I hope that my remarks will provide you with a good platform for your discussions during the rest of the conference.

In my speech today, I would like to talk about banks and banking after the end of the extended period of historically low interest rates which we have come to know as “low for long”. I will first consider the current state of the European banking sector, by which I mean those banks that are directly supervised by the ECB. I will then go on to discuss the potential lessons to be learnt from the market turmoil which we saw on both sides of the Atlantic in the spring of this year. I will also outline a few refinements to the supervisory framework which could be warranted as a result. Lastly, looking beyond recent events, I will highlight the challenges which banks need to tackle over the medium term – and hence also the issues which top the agenda for us supervisors.

Over the course of my remarks, I would like to convey three main messages.

The first is that the reforms to underpin banking activity enacted by global policymakers and regulators in the Basel setting in the wake of the global financial crisis have proven their worth. They have enabled banks to withstand the test of large and unexpected external shocks in recent years – first with the outbreak of the coronavirus (COVID-19) pandemic, and then with the fallout stemming from Russia’s war in Ukraine. Better regulation and more efficient supervision have thus helped to deliver a more resilient banking sector.

The second message has to do with the challenges posed by a rapidly changing monetary policy and interest rate environment. The global banking sector as a whole has benefited from the normalisation of monetary policy by major central banks, with higher net interest margins boosting bank earnings. We should not, however, lose sight of the fact that this trend is producing relative winners and losers among banks, partly on account of their size and their business models. More fundamentally, rising interest rates per se carry a number of risks which need to be appropriately managed by banks, as shown by the market turmoil this spring triggered by the failure of Silicon Valley Bank in the United States. So while the framework underpinning banking activity is not in need of major reform, in the current environment banks need to pay close attention to credit, valuation and liquidity risks. Adjustments in both reporting frequencies and the definition of certain liquidity requirements should be considered in order to help prudential authorities monitor risks to banks. Beyond the supervisory remit, it would also seem essential to enhance the transparency of credit default swap (CDS) markets.

Moreover, it is important to realise that weaknesses on the risk management side often go hand-in-hand with deficiencies in banks’ governance frameworks. This common feature of the banking failures which we saw in the United States and Switzerland this spring has also been at the core of several banking crises in the past. This is why I believe that, going forward, bankers and supervisors alike should continue to focus on both bank governance and risk management.

The third message is that, while in the near term we collectively grapple with traditional risks such as credit risk and market risk, which have become more prominent with the turning of the interest rate cycle, we must also prepare to address the emerging risks and challenges to the evolving banking landscape which will feature further down the line. We should apply a degree of caution when making our assessments in this regard, because the transition towards the steady-state of banking activity under monetary policy normalisation is still ongoing and the outlook remains subject to a measure of uncertainty. However, it is all but certain that, regardless of their business model, banks will have to make progress in preparing for the green transition and addressing the challenges posed by digitalisation.

Let me dive deeper into each of these messages in turn, starting with the current state of health of the European banking sector.

A resilient banking sector

If we took a snapshot of the key performance metrics for banks directly supervised by the ECB, we would come to the conclusion that the banking sector remains robust overall, with adequate capital and liquidity buffers, solid asset quality and improving profitability. In this regard, significant banks had an aggregate Common Equity Tier 1 (CET1) ratio of 15.5% in the first quarter of this year. This reading has remained broadly stable over a number of quarters now, at a level close to its all-time high. Banks appear to have an ample liquidity cushion, with the liquidity coverage ratio standing at 161% during the same quarter, well above both regulatory requirements and pre-pandemic levels. Seen in the round, asset quality is better than ever, with significant banks supervised by the ECB enjoying a headline non-performing loan ratio of 1.8% during the first quarter of this year, equalling the all-time low recorded in the third quarter of 2022. Profitability has been steadily improving in recent quarters, supported by the normalisation of monetary policy, with banks’ return on equity edging up to 9.6% in the first quarter of this year. This marks the highest point since the ECB took over direct supervision of the system in late 2014. As you know, lacklustre profitability has long been considered the Achilles heel of the European banking sector, so the fact that a growing number of the banks we supervise are in a position to achieve returns above the cost of capital in the coming quarters and beyond is a welcome development.

The robust position of the European banking sector as a whole was also confirmed by the results of the 2023 stress test, which the ECB published in late July.^[2] This showed that banks could withstand a severe economic downturn. In particular, the CET1 ratio of the 98 stress-tested banks would fall by 4.8 percentage points on average, to 10.4%, if they were exposed to three years of stress under a very adverse scenario assuming a prolonged period of low growth, elevated interest rates and high inflation. The negative impact on capital in the adverse scenario was driven by credit risk and market risk, coupled with lower income generation by banks. Loan losses generated 4.5 percentage points of CET1 ratio depletion, with unsecured retail portfolios being the most vulnerable. Another 1.4 percentage points of total capital depletion can be attributed to market risk in general, and to revaluation effects stemming from positions measured at fair value in particular. That said, a key result from this exercise was that capital depletion at the end of the three-year horizon (in 2025) was lower than in previous stress tests. This is mainly due to banks being in better shape overall going into the exercise, with higher-quality assets and stronger profitability.

Overall, we can safely say that the European banking sector remains on a firm footing, having successfully weathered the challenges brought about first by the pandemic and, more recently, by Russia’s war in Ukraine. These events have also seriously tested the Basel regulatory reforms. I think we can be pleased with the results, in that the framework has largely worked as intended. Looking ahead, however, we need to be aware that there are still some clouds on the horizon. In this regard, the second of the major external shocks to have hit our economies in recent years (Russia’s war in Ukraine) is still ongoing. This means that there is still an element of uncertainty to the outlook, even if this is currently less the case than it was a few months ago. Compared with the June 2023 forecasting round, the September ECB staff macroeconomic projections for the euro area contain a downward revision to the outlook for GDP growth – of 0.2 percentage points in 2023 and 0.5 percentage points in 2024 to 0.7% and 1% respectively – including on account of tighter financing conditions.^[3] Similarly, the ECB’s latest euro area bank lending survey^[4] also shows that credit supply conditions have tightened significantly since the turn of the year and loan growth has decreased sharply.

Insofar as the banking sector is affected, this implies that the positive contribution of interest margins to bank profitability is likely to fade to some extent, as higher interest rates are passed through to depositors and some downside risks materialise. In the first quarter of this year, the rise in banks’ funding costs exceeded the increase in their net interest income. Coupled with a weaker growth outlook, this also means that some deterioration in banks’ asset quality is to be expected in the months ahead. In this regard, while both the headline non-performing loan ratio and the Stage 2 loans ratio have decreased further in recent months, early arrears (meaning arrears between 30 and 90 days past due) have increased across the board.^[5]

As we saw in the spring of this year following the failure of Silicon Valley Bank in the United States, financial market sentiment remains fickle in an environment of tightening financial and credit conditions.^[6] Banks may thus be affected by sudden shifts in financial agents’ perceptions about their management risks in this phase of the cycle, especially as regards interest-related, valuation and liquidity risks. Let me now turn to the lessons that could be learnt from the episode of financial market turmoil earlier this year.

Lessons from the market turmoil in the spring

At the outset, I would like to underline that, from the ECB’s perspective, we don’t see any direct read-across from the bank failures which took place in the United States during the spring of this year to the banks under our supervision. There are two main reasons for this.

The first is that the banks we supervise do not share the same business model features as the regional banks that came under duress in the United States. This relates especially to the combination of high dependence on volatile funding sources – notably uninsured deposits – and material unrealised losses in securities portfolios held at amortised cost, which was characteristic of the troubled banks in the United States. By contrast, the banks we supervise tend to rely on a more diversified deposit base and do not exhibit extreme exposures to interest rate risk.

The second reason is that the perimeter of regulation for banks in Europe also differs from that in the United States. For example, regional banks in the United States have the option not to reflect unrealised losses^[7] on their available-for-sale assets in capital, but this is not possible for banks in the European Union. As you may know, the recent proposal by US federal banking agencies on how to implement the outstanding capital rules of the Basel III framework has moved to correct this discrepancy for US banks with total assets in excess of USD 100 billion.^[8] In addition, all EU banks are subject to the liquidity coverage ratio, whereas the troubled banks in the United States did not have this prudential requirement in place. Taken together, these differences help to explain why the median impact of banks’ unrealised losses in terms of capital is much higher in the United States than in Europe.^[9] The need to avoid our banks being exposed to the same weaknesses which we have seen in other jurisdictions is one reason why we at the ECB do not favour a loosening of our rulebook compared with the Basel III standards.

With this caveat in mind, the main lesson which both banks and their supervisors could learn from the period of financial market turbulence this spring is the need to redouble efforts to monitor interest rate and funding risks at a time when the monetary policy cycle is rapidly adjusting. These risks were already on our supervisory radar before the events in the spring. Indeed, our supervisory priorities in 2022 included interest rate and credit spread risk in the banking book, counterparty credit risk and the credit risk arising in segments of the lending business such as leveraged finance and commercial real estate. The 2023 stress test was also a helpful tool in this regard. This is because its results showed that banks’ ability to generate net interest income under an adverse scenario of increasing interest rates largely depended on their business model and the related asset and liability structure. For example, banks with a larger share of variable-rate loans benefited more from rising interest rates than those relying mostly on fixed-rate loans. Consequently, we have asked banks to pay even closer attention to this point.

In addition, following the events of the spring, we increased our scrutiny of banks’ unrealised losses. The related data collection exercise^[10] revealed that banks directly supervised by the ECB have around €70 billion of unrealised losses, net of hedging, on debt securities held at amortised cost. This figure is not material in aggregate terms, with US banks reporting USD 620 billion.^[11] However, although the banks we supervise do not have the extreme business model features exhibited by Silicon Valley Bank, such losses could be problematic when combined with weaknesses in banks’ asset and liability management. Separately, we have also been looking at exposures that are more sensitive to the turning of the interest rate cycle, such as commercial and residential real estate. We completed targeted reviews of banks’ related risk management practices earlier this year and are providing concrete recommendations to banks in this area.

The rise in interest rates also has implications for the funding and liquidity risks that banks may face. As it was long known that the ECB’s targeted longer-term refinancing operations (TLTROs) would fall due, we asked banks to prepare for this well in advance and to take into account the increase in market funding costs. The findings of our targeted review in this area show that although banks have an exit strategy in place to replace their TLTRO funding, some may face challenges in the current financial market environment. This review will be complemented by an analysis of banks’ liquidity and funding plans later this year.

Refining the risk monitoring framework

We know that interest rate and funding risks need to be closely monitored in a period of monetary policy normalisation, so the next question which we should ask ourselves is whether the framework in place to monitor banking risks needs to be adjusted to make this easier to achieve. The financial market turmoil we saw this spring has triggered a debate among policymakers and market practitioners on this issue, with both the Basel Committee on Banking Supervision and the Financial Stability Board set to assess what can be learnt from this episode.^[12] In our view, this is a very worthwhile discussion, and one in which the ECB will actively participate. However, policymakers sometimes have a tendency to jump the gun and enact regulatory reforms in the wake of a crisis. We think it would be better to hold back this time around. As I said before, the overall framework underpinning the activity of the banks we supervise has proven to be robust, including during the financial market turmoil in the spring. The fact that this episode led to a number of bank failures on both sides of the Atlantic does not necessarily mean that the framework is in need of a major overhaul, or that it should be recalibrated based on the deficiencies of the failed banks.

And even when the diagnosis of the problem is clear, the policy prescriptions which derive from it may be less straightforward. Let me give an example. Supervisory authorities in both the United States and Switzerland noted that the speed of the deposit outflows in the cases of Silicon Valley Bank and Credit Suisse was much greater than expected, especially among the uninsured deposits of non-financial corporations and financial institutions. While some observers have pointed to the influence of social media and online banking as contributing factors, finance professionals from such non-financial corporations and financial institutions do not appear at first glance to be the type to suddenly withdraw their funds with a quick swipe of their smartphones. And since there will always be an uninsured depositor base (unless a blanket guarantee were to be extended to the entire system, which would not be advisable owing to moral hazard concerns), then a potential increase in the threshold of insured deposits may not in itself help to prevent similar deposit withdrawal dynamics in the future.

Therefore, rather than reflecting a structural problem with the financial safety nets for bank customers in the information age or with online banking itself, the failures of Silicon Valley Bank and Credit Suisse may instead point to something much more fundamental. Once lost, confidence in banks exhibiting severe weaknesses in their business model and governance is very difficult to restore. This is why good governance and sound risk management by banks is key as a first line of defence to guard against sudden shifts in market sentiment. This is also echoed in the report produced by the US supervisory authorities following the demise of Silicon Valley Bank,^[13] which finds that the bank’s Board of Directors and management failed to manage their risks. The report also notes that supervisors did not fully appreciate the extent of the vulnerabilities as Silicon Valley Bank grew in size and complexity, and that even when vulnerabilities were correctly identified, supervisors did not take sufficiently prompt action to ensure that such problems were fixed in a timely manner. Taken together, the findings underline the need for bank managers to have sufficient capabilities to identify, measure and understand the risks inherent in their choice of business model as well as the importance of supervisors acting quickly in the face of uncertainty, provided they are sufficiently empowered to do so.

At the same time, however, the speed at which information was shared and affected decision-making by individual depositors and other market players during the market turmoil in the spring suggests that certain elements of the framework used to monitor risks to banks may need to be fine-tuned. One element has to do with how frequently and promptly data are submitted to supervisors. The metrics used to gauge liquidity requirements under the Basel framework, such as the liquidity coverage ratio and the net stable funding ratio, are available on a monthly basis at best, with a lag of a few weeks. While these indicators are still useful for comparative analysis, the events in the spring revealed the clearly limited ability of these indicators to flag issues in fast-moving situations where deposits are being withdrawn quickly. We have long been using our own metrics, such as the counterbalancing capacity,^[14] to assess banks’ preparedness to face sudden liquidity shocks. Now, we are looking to upgrade our analytical capabilities in this area by asking banks to send us their liquidity data on a weekly basis rather than on a monthly basis as before. This measure, which will be implemented in late September 2023, will be in place pending the European Banking Authority’s review of the frequency of supervisory reporting.

Another topic to consider is the extent to which bonds held at amortised cost can continue to qualify as high-quality liquid assets for the purposes of banks’ liquidity coverage ratios. From a prudential perspective, it would make better sense for banks to hold such bonds at market value, because they may find it hard to sell assets held at amortised cost without suffering losses when liquidity needs arise. Beyond the pure supervisory remit, the events of this spring suggest that the transparency of CDS trading also needs to be enhanced. This is because, although CDS markets are relatively shallow and illiquid, corporate and institutional depositors may use CDS spreads as triggers for withdrawals. All things considered, this implies that sharp price changes in these markets have the potential to destabilise banks.

Addressing climate-related risks and digitalisation

Looking beyond the events of recent months and the associated risks in the near term, as supervisors we also need to maintain our focus on the medium-term challenges that banks are facing in this constantly evolving landscape. Chief among these is making sure banks are prepared to embrace the green transition and address the risks stemming from the digital transformation.

We at the ECB do not regard climate change as a long-term risk, mainly because its impact is already visible and is expected to grow materially in the future. With this in mind, in recent years we have taken several steps to include climate-related and environmental (C&E) risks in our ongoing supervision.^[15] In its role as prudential supervisor, the ECB has made it clear that it is not in the business of telling banks how green their lending policies should be. However, it has also underlined that failing to take into account the transition towards a more sustainable economy would be incompatible with sound risk management. This is why we are insisting that the banks under our supervision manage C&E risks in the future in the same way as they would manage any other material risk today.

In this regard, the results of our 2022 thematic review on C&E risks show that, while banks have made meaningful progress in accounting for and addressing C&E risks, they still have some way to go to properly incorporate these risks in their risk management frameworks.^[16] Therefore, the ECB has set intermediate and bank-specific deadlines for achieving full alignment with its supervisory expectations in the area of C&E risks.^[17] We have asked banks to be fully compliant with these expectations by the end of 2024. If needed, we will step up the supervisory escalation to ensure compliance, which may include imposing periodic penalty payments or requiring bank-specific capital add-ons.

As regards digitalisation, our aim from a prudential perspective is to ensure that banks properly manage the long-term sustainability of their business model and the risks stemming from the digital transformation. As supervisors, we believe that banks can thrive on the opportunities offered by the digital transformation. But to do so, they need to prove capable of properly facing challenges such as strategic and execution risks, technology-related and operational risks and potential new emerging threats. Such risks need to be identified and addressed via governance and risk management frameworks and investment in capabilities.

With this goal in mind, we conducted an extensive survey to better understand the state of play in the digital transformation and the use of innovative technologies by the banks we supervise. The results, which we published earlier this year,^[18] show that while almost all banks have a digital transformation strategy, the degree of maturity differs. Most digitalisation strategies focus on improving the customer experience and offering digital services and products. And while a fifth of banks’ IT budget is spent on digitalisation on average, most banks do not yet have a dedicated digital transformation budget. The results also show that the banks themselves recognise that setting the right tone from the top down and having an effective internal control framework are crucial to enable the digital transformation. We are now following-up on these results via targeted reviews and on-site inspections^[19] on digital transformation, which will help to shape supervisory expectations in the coming years.

Conclusion

Let me conclude. On the whole, recent developments underline the importance of having well-capitalised banks, prudent supervisors, strong institutions and – to return to the words of Hartley Withers – good bankers, too. It appears that lessons have been learnt from the financial excesses seen in the run-up to the global financial crisis. Thanks to the collective efforts of all involved, the European banking sector as a whole has coped well with the many challenges it has faced in recent years.

However, this should not lull us into a false sense of security. The financial market turmoil we saw this last spring was a reminder that the path from “low for long” to monetary policy normalisation can be a bumpy one. More fundamentally, we know from our experience with crisis management that past performance is no guarantee of future results. This means we need to continue to challenge banks’ business models and risk management capabilities to ensure that they remain viable. While we will not shy away from this core supervisory task, we recognise that regulatory and institutional advances are also crucial to further buttress the resilience of the European banking sector. This is why the ECB is calling for the prompt implementation of the Basel III standards and the completion of the European banking union project.