Interview with Financial Times

Interview with Elizabeth McCaul, Member of the Supervisory Board of the ECB, conducted by Martin Arnold and Costas Mourselas

10 July 2024

What is the state of the Eurozone banking system right now?

I'm pleased with the state of the European banking sector. My view comes from the hard evidence earned in the school of hard knocks, where the banks have proven quite resilient in the face of some very significant challenges in the last few years. Looking back to the birth of the Single Supervisory Mechanism (SSM) ten years ago – when the banking system faced collapse or collapsed in different countries – it is simply remarkable how far the landscape has progressed since that period of time. It's very impressive what has happened here in Europe with the success of the SSM. We owe much to the visionaries who conceived it and to my predecessors who built the strong banking system we now have.

In 2015 the core capital ratio stood at 12.7% and now it has increased to stand at 15.7% in the first quarter of 2024. Non-performing loans (NPLs) were 8% in 2014, or €1 trillion; now, NPLs are 2.3%, or €355 billion. The banking system has continued to demonstrate resilience and garner confidence against a number of challenges, including in the face of Russia's aggression in Ukraine, and during the pandemic when the economy stalled, ways of working were transformed, supply chains were disrupted, and inflation and record-fast rising interest rates changed the market dynamics dramatically after such a long period of low-for-long interest rates. European banks also fared extremely well during the fateful events of March 2023 when other countries you wouldn't have expected had failing banks and very threatened institutions.

So, the first ten years of the SSM delivering on a strong euro area banking system is an extraordinary story. Having said all that, no supervisor is ever complacent, and I continue to have a cautious stance given the significant uncertainties that still exist.

What are the main risks and concerns you can see? If you had warning lights on your desk, which ones would be flashing amber or red?

We are always looking for the warning lights on the dashboard for disturbances which may be lurking just beneath the surface of deceptively calm waters. I wouldn't characterise our posture as sanguine at the moment. There are certainly caution lights in front of us. For me the most prevalent one is the area into which we likely have the least visibility and where things can move faster than, for example, the normal credit dynamics. That is the non-bank financial intermediaries (NBFI) market, which has doubled in recent times: in the euro area in 2008 it was €15 trillion, growing to €32 trillion in 2024. Globally, the NBFI growth number is even more worrying, increasing from €87 trillion in 2008 to €200 trillion in 2022.

We are placing particular focus on the private equity and private credit markets. The growth there is especially noteworthy. In 2012 it was globally €3 trillion in the private equity markets. And in 2024 it's €8 trillion. The European private credit market has grown 29% in the last three years. At the end of 2023 the euro area private credit market accounted for about €100 billion, or 6% of the global market. Globally, it is now above €1.6 trillion.

This growth is remarkable and something that always worries us. And it is outside of the banking supervisory and regulatory perimeter. I have been involved in supervision in one way or another for a long period of time. You learn your lessons on the job. I was at the table when Long-Term Capital Management (LTCM) collapsed, when we learned the perils of correlation risk generated by same direction trading strategies on the books of a hedge fund to positions taken by banks we supervised. When LTCM’s trading strategy went south, it had an exponential impact on the banks, requiring a private bailout to avoid the implications of what was then considered an enormous systemic factor in the overall market. I suspect correlation risk is occurring again. We've had blips. Maybe even more than blips. Archegos was certainly a warning light. The gilt market dislocation caused by liability-driven investments in pension funds in the United Kingdom was another warning light, and we had hedge funds going down during the COVID crisis.

We know risk from the NBFI market can crystallise in several ways. It can be via the correlation of exposures where, especially given the growth in the private credit and private equity markets, maybe we have the same exposures as ones on bank balance sheets and where risk can spill over in various ways. It may also be the case that there are hedging strategies to the same exposures in both the regulated banking market and the NBFI market. And finally some of these funds, especially certain hedge funds, are becoming so big that they can partially move the market by themselves and are not likely to act as shock absorbers in the same way banks have sometimes acted. This matters also for liquidity dynamics in the system.

We don't have a fully clear line of sight from the banking supervisory standpoint about the level of exposures that may be correlated to risk on NBFI balance sheets from banking industry lending arrangements, including repurchase agreements, lines of credit or derivatives to and from NBFIs.

If you talk to me for five minutes, you will hear me ringing the bell about my concerns. I worry about how we are getting our arms around our understanding of the risk emanating from this sector of the market to the supervised banks. This is because of the opacity of the NBFI sector and the resulting lack of understanding preventing us from developing a view about exactly what the interconnectedness translates into for systemic risk.

So what are you doing about this?

There are a number of international projects under way. One of the most important ones is work currently under way at the Financial Stability Board (FSB), which is looking to identify best practices, policies and procedures that should be adopted by a wide variety of supervisors that have at least some oversight in this area. What is required is to develop approaches to addressing some serious questions in a practical way. How do we get the right data? What kind of data reporting should be required, by whom, and to whom should it be delivered so that that picture can be really filled in in an appropriate way? Meanwhile, one focus area of the ECB and other supervisors is counterparty credit risk. If an institution has lending arrangements, trading arrangements or hedging strategies connected to the NBFI market, we are asking what line of sight and due diligence they are conducting. They need to have a holistic risk picture about how a change in the status of a counterparty could reverberate onto their balance sheet in some way if they are providing credit or conducting trading with the counterparty.

Do you think that Archegos could have been avoided if you’d had this more holistic picture?

Let me be clear that the banks themselves are the first line of defence. In the first instance, what I hope is that the industry has access to the information necessary to enable it to make appropriately risk-based decisions about the use of its balance sheets and how banks are providing credit lines or derivative products, how they are interacting with the counterparty. Archegos is an excellent example of the dangers of opacity. It wasn't only that there were risky exposures on the books of Archegos, they were entering into lending arrangements with multiple institutions where the institutions didn't have appropriate covenants in place.

In the second instance, and maybe equally as important, the supervisors need to have access to more data. The conversations that are happening in a wide variety of supervisory authorities about gaining visibility into the whole picture are very worthy discussions.

But it feels like you've been talking about this for a long time, over a decade. What's holding this up?

I’m happy to see there are discussions at the Financial Stability Oversight Council (FSOC) about the contours of the regulatory perimeter. There are discussions at the Prudential Regulation Authority (PRA) on the regulatory perimeter. There is global work at the FSB. These are serious efforts. Perhaps it is indicative of the global nature of the markets that this picture needs to be put together by more than a single regulatory regime. This is something that's going to require a concerted effort internationally across the board. At the ECB, we're doing everything we can to communicate what banks have to do with counterparty credit risk that is connected to this part of the market. And we are working together with our colleagues, for example in the PRA and the Federal Reserve System, to get a better holistic view of the risks for these global players.

You mentioned FSOC and the recent changes in the United States to come up with a framework to identify these really important non-bank institutions. Does Europe need a similar framework to identify some of these behemoths that might need more supervision?

We don't have an FSOC but we do have the European Systemic Risk Board (ESRB), which is focused on identifying a view on broader segments of the financial sector and identifying interconnected risk. So, I wouldn't see the need to call for an FSOC-like entity since we have the ESRB. But we need better and appropriate data also for supervisors and macroprudential authorities and we should also think further about European supervisory powers here.

Do you also have concerns about leveraged lending within the banks?

Leveraged lending has been on our minds at the ECB since 2017, when we issued very strong guidance to the banks. Certainly, the pandemic brought into sharper focus what was happening with the leveraged lending sector in the supervised banks, largely due to concerns about the potential for increases in defaults and in pipeline losses. We have utilised a variety of the spectrum of supervisory processes available to us – horizontal reviews, targeted reviews, Dear CEO letters – to highlight issues to institutions. We have published best practices and seen improvements, especially in risk controls for leveraged lending. And we have taken strong supervisory actions in a number of cases. But there are also areas that continue to give us concern, such as how banks are assessing credit risk and staging and provisioning. So we are still quite active on this aspect.

After the turmoil in the banking sector in March 2023, it seems that one of the big conclusions of global regulators and supervisors was that they should not wait to have the full information before acting on an issue they have identified. Are you at the SSM now quicker to take action rather than letting these things fester?

In the summer of 2022 Andrea Enria, Chair of the Supervisory Board at that time, had a crystal ball. He called for an outside look at the effectiveness of our supervisory process by a global, expert group of luminaries, people above reproach in the area of supervision, in light of the tenth anniversary of the SSM this year. We asked them to look at the annual Supervisory Review and Evaluation Process (SREP), which is at the heart of our banking supervisory process. The experts delivered their report in the week of the market dislocations caused by Silicon Valley Bank and First Republic Bank, followed shortly after by Credit Suisse. In the aftermath of the US bank stresses and the Credit Suisse sale to UBS, it seemed the whole world was asking about the effectiveness of supervision. The SSM instead had a report delivered that found our supervision to be effective.

The expert group made recommendations to enhance ECB Banking Supervision to make it fit for purpose for the next ten years. They recognised that a level playing field and recapitalising the European banking system were the first items of business for the SSM since it was created at the time that rebuilding confidence in the European banking system and breaking the sovereign “doom loop” were of paramount importance. As a result, the SSM’s supervision was necessarily geared in a “capital-centric” fashion, to quote the group.

They made strong recommendations about complementing the supervisory toolkit with other tools, especially in areas that don't lend themselves as easily to the use of a capital charge in the first instance for creating the conditions for improvements in a deficiency in the bank. Business model sustainability and the strength of the governance framework are good examples where it can sometimes be more effective to ask banks to develop and execute remediation with action plans with embedded, time-bound accountability requirements, and use other tools in the supervisory toolkit to escalate oversight and ultimately enforce remediation where it's not occurring. Some of the expert group recommendations are about enhancements to the supervisory process that the ECB itself was studying, and in some cases it already had implementation plans under way.

So you are showing your teeth more as supervisors?

When we put remediation requirements forward to a bank in the form of operational acts and as findings and measures, we want to be specific about our follow-through. We want to make certain that remediation has in fact occurred to standard. If not, we are being more deliberate about the necessary escalation steps we're going to take, for example in the form of legally-binding decisions with sanctions attached if banks do not deliver. We are very much moving in the direction of following the escalation ladder we adopted.

We are reforming the SREP process in a wide variety of ways. One is making fuller use of the supervisory toolkit to ensure that we have very clear linkages between the escalation ladder and the SREP process, and how we position our overall supervision to ensure we are following through on deficiencies in supervised institutions identified effectively.

This focus on escalation and greater scrutiny on follow-through on remediation is not about reducing capital requirements or not using capital requirements when needed. We use capital add-ons, for example, in the case of leveraged finance. Capital add-ons certainly can move the needle in terms of getting an institution to remediate to reduce risk.

But there are other tools that are more effective tools to use in the first instance. Governance is a good example, including getting the right management body structure in place, having the right expertise at that level, having the right metrics and requiring systems to have the right data so that appropriate risk analysis can be made. If important deficiencies in areas such as these are not being effectively remediated in a timely manner, then effective supervision means ultimately moving up the escalation ladder to obtain the right outcome. This includes putting a periodic penalty payment process in place and then using that process as part of our supervision.

We expect institutions to make effective remediation plans with milestones and meet deadlines. And if there is not timely remediation or deadlines for meeting milestones are being missed, then we need to follow through with appropriate escalation measures. We have not really used those muscles to a larger extent so far at the ECB. Periodic penalty payments are part of the toolkit delivered to us by the legislators in the SSM Regulation. The SSM has done an excellent job with its focus on the recapitalisation of banks and really making sure that a consistent approach has been delivered. Now it's time to also use additional tools, such as periodic penalty payments, when necessary.

You have started to impose periodic penalty payments on banks that don’t meet your expectations on climate risk – something you have never done before at the SSM. Is this a test case for other areas?

I think climate risk is in a category of its own. I wouldn't use the words test case. Beginning in 2020 we realised institutions had a great deal of catching up to do in order to onboard climate risk into their traditional risk metrics for credit risk and operational risk.

Initially, we asked banks to conduct a self-assessment. We followed up with a targeted review to understand whether the banks’ self-assessments were appropriate. We communicated to institutions where we thought there was a gap between their self-assessment and what we saw when we went on site. We conducted a stress test that gave us more visibility into the processes. We told banks very clearly in 2022 what expectations we had for closing gaps in basic areas, to have transition plans in place, to have a way of assessing credit risk in the portfolio emanating from climate risk, to have risk analytics in place. And we told banks to deliver on certain things by March 2023 and year-end 2023, and that addressing climate risk be in full alignment with our expectations by the end of 2024. Along the way extensions were also given in certain cases, because at the end of the day what is important is that climate risk is integrated into banks’ governance, strategy and credit and operational risk management. So there has been very, very clear up-front communication on both an individual basis with institutions that have gaps and, also more broadly, in our communications to the market.

I think it's a particular case where we will be following through where necessary, and communicating to banks that if the milestones are not met then we will make full use of the supervisory toolkit that's available to us. Of course, our hope is that they deliver on our expectations and the needle is moved in the institutions. That's the goal. But if it's not happening, we will move forward with the other tools that we have available in order to make certain those requirements be put in place.

Climate risk is also rather particular because I think in many ways we were coming from such a low level of implementation across the board. At first, it was nearly written off as a social policy, even though expectations were set forth to consider climate risk as part of credit risk and operational risk. Then it got moved into the risk area and was taken more seriously. There are other areas that are more long-standing. I would put risk data aggregation in that category.

What does “risk data aggregation” mean?

There is a big investment that needs to be made for banks to be able to assess risk and assess exposures. It's a costly undertaking, but it is utterly essential if at the end of the day our supervised institutions can identify, manage and mitigate risks to their balance sheets.

We are focused on this as a key ingredient for the efficacy of risk management and are calling attention to the need for improvements. We are requiring institutions to make the necessary investments. Risk data aggregation is an area that probably lends itself quite well to the use of other tools in the toolkit rather than, in the first instance, a capital charge.

Are you saying that you're planning to use periodic penalty payments for banks that are laggards on risk data aggregation?

That would be too strong a statement. Instead, I would say that in any area where institutions are lagging behind on following through on the remediation that we've required them to do, we will use the full supervisory toolkit to ensure that timely action is taken. And requiring timely remediation, so banks have a line of sight into and manage risk appropriately, is the definition of effective supervision.

On private credit, the argument that the industry makes is that this makes the system safer, because it's no longer the banks which are making as many loans to companies. That means less systemic risk. Do you agree with that broad argument?

No, not at all. I'm reminded of the sub-prime crisis. I was the supervisor in New York in the early 2000s. We had supervisory oversight over segments of the market that were not traditional banks, such as the mortgage banking and mortgage broker market. I have investment banking training, so I also understand securitisation.

We took a very strong stance at that time against predatory lending. I brought some of the first cases that involved sub-prime lending, where there were abusive practices taking place. Following those actions, we initially couldn't understand why the market supporting predatory lending practices continued to grow. How was it possible that the sector continued to make loans, including predatory loans? Well, the mortgage bankers had lines of credit to the banking sector and the ability to offload mortgage loans through securitisations structured by the investment banks.

The New York State Banking Department did not supervise investment banking activity in the United States. But securitisation prospectuses contained the information that there was a moral hazard. There was a requirement in the prospectus that the originator remained on the hook for replacing mortgage loans made outside of the disclosed deal compliance requirements, for example, for compliance breaches north of 5%. When the mortgage bankers lost their bank lines of credit and were unable to replace non-compliant loans in the securitisation, the securitisations went bust. It's exactly this picture that comes to mind when I think about how disintermediation is seen on the surface to remove risk from the banking sector. But the connection points between various sectors of the financial services industry mean that we need to be cognisant of the interconnectedness risks. We need to understand where the interconnectedness presents risks to the banking sector. And we should not forget that these credit funds have not so far been through a full credit cycle, some of them are not close-ended and subject to liquidity risk, and in the end there are pension and insurance funds as end investors here. So this is an important issue to follow for the future.

The ECB has been telling Eurozone banks for over two years that they need to exit Russia. But some big banks in Europe still have large positions in Russia. What can you do about this?

I wouldn’t characterise supervised institutions’ Russian exposures as large. The exposures in fact are quite contained overall in terms of the volume that we see. And there's been a downward trend, which we very much welcome. Exposure declined by approximately 55% between the onset of the war and the end of 2023. The institutions have been quite cognisant of the potential impact of potential losses coming from banks’ remaining exposures. We're obviously carefully looking at the capital and liquidity positions of those banks and they're overall robust, so we think that would be a mitigation. Having said that, we've asked banks to speed up their de-risking efforts by setting clear roadmaps towards substantially de-risking and to report regularly to their management bodies and to the supervisor. Based on the risk evolution and the progress of each bank, further supervisory actions can be considered to address the individual situations if they don’t follow through. We have a wide variety of tools available to us. The bottom line is that this is clearly a risk area where we have asked institutions to sharpen their focus.

If US regulators were to take enforcement action over one of these banks because of their Russian activities, that could be very serious, existential even, couldn’t it?

As you know, the ECB is not the anti-money laundering or the sanctions supervisor. But having said that, we could not be more aware of the nexus between risks emanating from sanctions in the anti-money laundering arena and prudential risks. There are plenty of examples that should be front of mind for institutions analysing their Russian exposures. Ultimately, this risk can be very important for an institution. Sanctions can represent a significant impact on the balance sheet. And it can even present franchise risk. We have seen institutions in the European market designated by US authorities as institutions of money laundering concern. And the story after such a designation is one of the franchise.

Commercial real estate prices are still falling in Europe – how worried are you about this area?

We continue to be in the midst of a societal and structural change with respect to commercial real estate (CRE), which started during the pandemic and is continuing. Added to that are other risks emblematic of commercial real estate, such as environmental risks. The higher interest rate environment has brought more profitability to banks. But we're very focused on the office arena in the United States and the need for refinancing. CREs have bullet maturities in the US that require significant refinancing over the next couple of years, and the higher interest rate environment coupled with low office space occupancy and leases not yet rolled over provide significant challenges. I know that the European banks, as a general matter, are much less exposed to the US sector overall and on the whole I would say this is not a systemic risk. But where those exposures exist, it’s necessary to have appropriate risk management in place. And lastly, the commercial real estate market isn't only financed by the traditional banking sector. So, where there are interconnections to the NBFI market, this has to be well understood.