Securitisation: you can never tranche the same portfolio twice

Keynote speech by Pedro Machado, Member of the Supervisory Board of the ECB, at the European Financial Institutions Conference

Frankfurt am Main, 30 September 2025

It is a pleasure to be here today to address a topic which is certainly of great interest for this audience and S&P as a leading rating agency, and which is at the top of the EU’s regulatory agenda: securitisation.^[1]

The title of this speech is inspired by a quote about change from the Greek philosopher Heraclitus: “You can never step into the same river twice”. It is a not-so-veiled reference to the EU ban on re-securitisation. But that is not the only thing that has changed for EU securitisation markets since the global financial crisis.

Securitisation is a very polarising topic among industry participants and policymakers.

Today I would like to provide a neutral and fact-based perspective on securitisation – by considering its potential benefits and risks for the EU, examining the lessons from the global financial crisis, highlighting more recent market developments in EU securitisation markets, and looking at what we can expect in the future.

Let’s start with some basic concepts: to use a very simple definition, from a regulatory point of view, securitisation is a financial technique whereby the credit risk of a loan portfolio is tranched into notes of different seniorities, which can then be transferred to other investors or retained by the originator.

Securitisation in itself is not intrinsically good or bad – it is a powerful tool that can benefit those who use it, and potentially also the broader economy. But it does involve some risks that are inherent to the technique itself.

Looking at the benefits of securitisation one by one, we can start with the originator’s perspective: the bank issuing a securitisation can transfer part of the credit risk associated with the securitised portfolio. This is particularly useful in managing risks, for instance to limit excessive concentration to certain sectors or individual borrowers. From a regulatory perspective, where the transfer of risk is deemed significant, the regulatory framework allows capital requirements on the securitised exposures to be reduced, so that the bank’s overall requirements more closely reflect the risks it actually retains.

Investors also benefit from securitisation – by accessing credit risk instruments which allow them to diversify their investment portfolio and satisfy their risk appetite.

Securitisation, under certain conditions, can also improve the financing for the real economy. By transferring risks to a broader investor base, it can create space on banks’ balance sheets for new business, depending on their strategy.

In addition, securitisation contributes to the development of the capital markets union thanks to its unique ability to act as a bridge between banks and capital markets: the risks and rewards in the form of securitisation tranches can be shared across many financial investors, also on a cross-border basis.

To sum up: for banks, securitisation allows part of the credit risk of a portfolio to be transferred, reducing sectoral or borrower concentration and, where risk transfer is deemed significant, lowering the associated regulatory capital requirements so that they better reflect the risks the bank actually retains. For investors, securitisation provides access to diversified exposures and instruments tailored to different risk appetites. And for the real economy, securitisation can create space on banks’ balance sheets for new business, depending on their strategy. It also supports the capital markets union by connecting banks and markets, allowing risks and returns to be shared across a wider pool of financial investors, often across borders.

But this is also where the risks may come from. US securitisation markets were one of the main transmission channels that allowed the crisis in the US real estate market to propagate to the rest of the world.

In the years following the global financial crisis, the G20 and global standard-setting bodies such as the Basel Committee addressed the main shortcomings that had been identified in the regulatory framework.^[2] First, the weak alignment of interests between originators and investors, which materialised as an originate-to-distribute model. This has now been mitigated by new risk retention requirements for originators, to avoid relaxing origination and servicing standards. Second, deficiencies in how banks calculate the capital they have to hold against retained securitisations: this was addressed in the new securitisation standards adopted by the Basel Committee. Third, the asymmetry of information between originators and investors, which was addressed through enhanced disclosure requirements for securitisation and alignment of credit granting criteria for securitised and non-securitised exposures. Fourth – and this one will certainly not come as a surprise to anyone here – the overreliance on external credit ratings was also an amplifier of the crisis, as investors neglected their own due diligence duties and invested blindly in complex structured products, such as the CDO squared, which have now been banned almost completely in the EU.

Overall, the lessons from the global financial crisis have been largely learnt, particularly in the EU, where the Securitisation Regulation adopted in 2017 and the concurrent implementation of Basel III standards considerably enhanced the regulatory framework. We can therefore safely say that securitisations issued in the EU today are considerably safer than they used to be before the global financial crisis.

That being said, even accounting for the sizeable layer of additional safety provided by the new rules, securitisation retains an intrinsic level of risk that is inherent to its nature as a tool designed to pool, tranche and distribute the credit risk of a portfolio of individual loans. For example, the impact of small errors in the estimation of loss parameters, which would ordinarily be limited for an individual loan, can be exacerbated by the tranching of risks. Investors could also fall for the illusion of safety in senior tranches of securitisations, which often enjoy the highest possible credit quality for structured products, while remaining exposed to systemic risk.

As is often the case with regulation and supervision, it is crucial for all of us to make sure that we are not fighting yesterday’s war, and that we do not fall for the preconceived idea that securitisation markets have to look the same.

If we want to have a regulatory framework that is really fit for purpose, we need to understand how the EU securitisation market has developed over the past few years.

I often come across comparisons between the EU and the US securitisation markets, and many stakeholders seem to think that the EU lags behind the United States. But I disagree, on two main grounds. First of all, these comparisons often fail to account for the fact that a significant share of the US securitisation market comes from the US government agencies. Actually, most of the securities issued by the agencies do not involve tranching of risk and are therefore not treated as securitisations from a prudential perspective. Also, they rely primarily on the implicit or explicit guarantee of the US Government, which reflects a public policy choice that differs from the European approach. And second, when excluding issuances from US agencies, and considering only outstanding traditional securitisations, the difference in volumes between the EU and the United States is roughly one to three^[3]. However, this does not capture the rapid growth of synthetic securitisations in the EU.

What are synthetic securitisations? In essence, they are a guarantee on one or several tranches of credit risk from a loan portfolio. Investors receive a fee for this protection, while the securitised loans remain on the balance sheet of the originator. As such, synthetic securitisations do not provide funding to the originator. They are not usually traded on a secondary market, although they can still be transferred to new investors in the form of credit-linked notes. And – sorry to highlight this at an S&P event – they are often not externally rated.

As mentioned, synthetic securitisations are a very effective tool for banks to transfer credit risk to third parties. Where the supervisor – in many cases the ECB – is satisfied that the transfer is significant, the regulatory framework allows the bank’s capital requirements to be reduced accordingly, so that they more closely reflect the risks it continues to retain.

In practice, this means the ECB needs to assess whether individual securitisations achieve significant risk transfer or not. In addition to this transaction-level assessment, the ECB scrutinises banks’ risk management frameworks to ensure they are sufficiently well equipped to originate sound risk transfer securitisations. We are also mindful that banks should not rely excessively on securitisations for their capital planning.

The growth of the EU synthetic market is already well documented, for example thanks to the work of the European Systemic Risk Board.^[4] As at the end of 2023, the EU was the largest market for synthetic securitisations in the world, representing half of the global outstanding amount, while the United States accounted for only 20%.

The volume of synthetic securitisations issued by significant institutions has grown markedly over the last few years, increasing by 24% between 2022 and 2024. For the first half of 2025 alone, we observe an even stronger increase – slightly above 85% – compared with the same period in 2024. This is also consistent with the continuous rise in transactions aimed at achieving significant risk transfer.

It is important to understand the drivers of this rapid growth.

First, synthetic securitisations are cheaper to issue than traditional ones and, importantly, although they do not bring funding to the originator, they can transfer a significant share of credit risk, Where the transfer is deemed significant, the regulatory framework allows lower capital requirements on the securitised exposures. Banks can therefore combine the risk-transfer benefits of synthetic securitisations with cheaper sources of funding, for instance in the form of covered bonds.

Second, while significant institutions have used synthetic securitisations on a regular basis for several years now, they were previously reliant on a limited number of long-standing investors. Recently, new investors have joined the market, which has facilitated the growth in new issuances.

And third, the greater number of investors has also improved market conditions for originators, making synthetic securitisation even more attractive for them.

When securitisation achieves significant risk transfer, capital requirements are adjusted to reflect more accurately the risks a bank retains. How banks use their capital position thereafter is a business choice, and from a supervisory perspective there is no automatic link to additional lending.

Securitisations can also support financing more directly, for example through asset-backed commercial paper programmes, which support companies in financing their daily operations.

We can see that the EU securitisation market has developed a strong preference for synthetic securitisation, which reflects the specificities of the European financial sector. These developments, rather than the benchmark of the US market, should drive our rulemaking process. Otherwise, we may end up being disappointed by its outcome, and we risk overlooking the actual threats to financial stability building up in front of us.

That brings me to the European Commission’s recent legislative proposals on the EU securitisation framework. While I will leave it to the upcoming ECB opinion to offer a detailed assessment of the proposals in due course, I want to share some initial thoughts with you.

First of all, I agree with the overarching objectives of the proposals. A well-functioning European securitisation market is an important element of the savings and investments union agenda, and a well-structured securitisation framework can support bank lending and, ultimately, economic growth, helping to maintain financial stability. That being said, we know there is no intrinsic link between securitisation and additional lending, which is why I welcome the Commission’s intention to monitor lending activities under the proposed new rules.

Second, and consistent with these overarching objectives, securitisation rules should support simpler, more standardised and more resilient transactions. This would incentivise new originators and investors to enter the market, facilitate their due diligence duties and ensure we avoid the kind of mistakes that led to the global financial crisis. Complex securitisations and opaque structures may maximise the economic efficiency of individual transactions in the short term, but their higher complexity puts a strain on the resources of originators, investors and supervisors. From my perspective, this is not what a sustainable securitisation market should look like.

Third, as a general principle, we should avoid unnecessary complexity in rules wherever possible, and the securitisation framework is no different. Requirements should be streamlined and made more proportionate to the extent possible without putting financial stability at risk. This is particularly the case for the proposed changes to due diligence, risk retention and transparency requirements.

Fourth, the proposals aim to strike a balance between reducing capital requirements and ensuring the resilience of the affected transactions, which is a sound approach. However, I fear that the proposals may have gone too far in reducing capital requirements for more complex transactions. The proposals would also significantly increase the complexity of the framework. We must remember that the practice of securitisation entails inherent risks to financial stability, such as agency and model risks, which are more difficult to capture in more complex securitisation structures. Against this background, I believe that the proposals could be simplified and adjusted to focus more on simpler, more resilient transactions.

Fifth, I welcome the amendments proposed in relation to the significant risk transfer assessment process. They would create a clearer, more up-to-date framework that is better aligned with the current practices of banks and supervisors.

And one final reflection: the proposals also acknowledge the efforts made by the ECB, together with the industry, to develop a new fast-track process for simple securitisations. This has been the flagship initiative of our broader agenda to simplify supervisory processes, which makes it all the more important to us. It will reduce the time to market for originators – the time needed for our assessment will be reduced from three months to eight working days – and it can also help new issuers understand our supervisory approach and expectations regarding significant risk transfer. All of this should, in turn, be conducive to a broader EU securitisation market. The success of this fast-track process will ultimately depend on the banks' willingness and ability to structure securitisation transactions in a sufficiently standardised manner. In other words: it’s also up to the banks to contribute to simplification by adopting sufficiently simple securitisation structures.

To conclude, securitisation has changed profoundly in Europe since the global financial crisis. It is safer, more transparent, and increasingly adapted to our financial system. But it remains a tool that should be used with caution. If we continue to ensure simplicity and transparency, securitisation can make a meaningful contribution to financing the European economy, while preserving financial stability.

Thank you.