Unlocking mortgage risk transfer: Why securitisation floor calibration and insurer access must move together

Abstract
EU bank-centred mortgages remain largely absent from significant risk transfer (SRT). This Note argues that the proposed Resilient label for simple, transparent and standardised (STS) securitisation can unlock mortgage SRT only if two conditions are met: a senior risk-weight floor low enough to make capital relief usable, and EU (re)insurers providing unfunded credit protection eligible for STS Resilient treatment. Using Pillar 3 disclosures from 40 EU-regulated banks with €4.5trn of mortgage exposures, we estimate that the current 10% STS RW floor leaves 6% in the viable perimeter; a 5% floor raises this to 18%; and a 2% floor to 50%. But a lower floor without insurer access would create a market-access trap: banks would issue long-term mortgage SRT securitisations in STS Resilient format, while insurer-supported transactions may sit outside this segment. To succeed, reform must combine a usable floor with coherent insurer access.

Europe’s missing mortgage risk-transfer market

EU banks are the main providers of credit to the EU housing market and hold the overwhelming majority of residential mortgages on their balance sheets. Covered bonds provide an efficient funding channel for these assets, but leave the underlying credit risk and associated regulatory capital requirement with the originating bank.1 This creates a structural gap: efficient mortgage funding, but only limited mortgage credit risk transfer. Significant risk transfer (SRT) securitisation can fill this gap by enabling banks to transfer part of a portfolio’s credit risk to investors or protection providers and, where supervisors recognise the transfer as significant, obtain regulatory capital relief. In a synthetic transaction, this risk is transferred without selling the underlying assets. Synthetic SRT transactions are already well established in Europe for loans to small and medium-sized enterprises and corporates, but remain marginal for residential mortgages. This leaves Europe’s largest bank balance-sheet asset class without a scalable risk-transfer market.

This matters because residential mortgages, despite their relatively low average risk-weight (RW) density, generate substantial risk-weighted assets (RWA) given their scale.2 Their low margins make the cost of retained capital particularly material for bank profitability and competitiveness. A scalable mortgage SRT market would allow banks to transfer targeted credit risk while preserving covered bond funding and optimising risk transfer for the parts of mortgage portfolios where capital relief is most valuable. The policy question is whether mortgage SRT can deliver enough capital relief to justify protection, structuring and supervisory execution costs.

The EU already provides favourable capital treatment for securitisations meeting the “simple, transparent and standardised” (STS) criteria, which are intended to make their structures and risks easier to assess. The June 2025 EU Securitisation Package proposes an additional “Resilient” category for senior securitisation positions meeting further structural safeguards, including requirements relating to portfolio granularity, amortisation and the senior tranche attachment point. Positions meeting both sets of criteria would qualify for the most favourable capital treatment as “STS Resilient”. For residential mortgages, however, this treatment will matter only if it improves transaction economics in practice. Achieving this requires both a senior risk-weight floor (RWF) low enough to permit proportionate capital relief and access to a broad, reliable base of long-term protection providers, including regulated (re)insurers3 providing unfunded credit protection (UFCP).

There is a specific policy trap to avoid. Several Resilient safeguards are already embedded in STS transactions, notably granularity and amortisation criteria. The main practical distinction between STS Resilient and STS Non-Resilient is therefore likely to be the senior attachment-point condition. Because that attachment point is largely within structuring discretion, mortgage SRT transactions are likely to be structured as Resilient whenever possible to maximise recognised capital relief.4 A lower STS Resilient RWF without (re)insurer access would make the economically relevant STS segment available primarily through eligible funded protection providers,5 while leaving insurer-supported STS transactions outside the economically usable STS segment. In practice, the reform could shift activity away from regulated insurance capacity precisely where long-term risk-bearing capacity is most needed.

Why the senior RWF calibration is decisive for mortgage SRT issuance

The proposed Securitisation Package would make the senior RWF more risk-sensitive to the RW of the securitised pool, but still subject to a minimum floor. For low-RW assets such as residential mortgages, that floor can bind, leaving the retained senior tranche with a disproportionate capital charge relative to the residual risk retained and undermining transaction economics.

A simple capital-relief RW threshold identifies the pool-average RW at which SRT becomes economically attractive. After the introduction of the synthetic STS framework, IACPM survey evidence for 2022-2024 places the lower tail of synthetic SRT pool RWs close to 40%, against an average of around 60%.6 With euro-area banks’ average RW density estimated at 33%,7 this 40% benchmark helps explain why large parts of bank balance sheets, including residential mortgages, sit outside the viable SRT perimeter. Under the current 10% STS senior RWF, it implies a practical 4.0x relationship between the floor and the pool-level RW threshold.

Applied as a transparent rule of thumb, not a transaction model, this gives pool RW thresholds of 20% for a 5% floor and 8% for a 2% floor. The point is to show how floors affect the mortgage exposures capable of supporting economically viable capital-relief securitisation, and to give policymakers a practical measure of the impact of their calibration decision.

The calibration of the minimum senior RWF therefore determines whether STS Resilient can create an effective and scalable EU mortgage SRT market, freeing up capacity for new lending, or only a formal legal possibility with limited benefit for the EU economy.

Evidence from 40 EU-regulated IRB banks’ mortgage portfolios

To assess the effect of alternative RWF calibrations, we use Pillar 3 CR6 disclosures from 40 EU-regulated IRB banks selected for mortgage-market relevance, data availability and plausible SRT participation. The sample covers major and medium-sized mortgage lenders headquartered in 13 European jurisdictions. It represents €4.47 trillion of residential mortgage exposure at default (EAD), €626 billion of residential mortgage RWA and an average RW of 14.0%, compared with an estimated EU average residential mortgage RW of around 17%.8

The aggregate RW distribution is wide, as shown in Table 1. Exposures with PDs of 2.5% or above, including defaulted exposures, account for 6.3% of aggregated EAD and have an average RW of 72.6%. By contrast, the STS-relevant mortgage perimeter used in the analysis is the non-defaulted and non-distressed segment with PDs below 2.5%, representing 93.7% of aggregated EAD and an average RW of 10.1%. This is where STS Resilient calibration matters most: these assets form the core candidate perimeter for mortgage SRT securitisations seeking STS treatment, but their low average RW makes capital relief highly sensitive to the retained senior floor.

Table 1. Characteristics of the aggregated residential mortgage non-SME portfolios of the 40 EU-regulated IRB banks
(by PD range, in € millions)

Because RWs vary both within and across PD buckets, Table 1 alone cannot determine the economically viable SRT perimeter. We therefore aggregate each bank’s cumulative RW distribution in Figure 1 to estimate viable mortgage EAD under alternative senior RWFs.

The orange series in Figure 1 plots the cumulative distribution of residential mortgage EAD by RW across the €4.47 trillion sample, using the left-hand axis. The dashed vertical black line separates the STS-relevant mortgage perimeter, namely the 93.7% of aggregated EAD with PDs below 2.5%, from the 6.3% high-PD and defaulted tail excluded from this analysis. The solid vertical threshold lines show the individual-exposure RW boundaries used to estimate the viable EAD perimeter under each senior RWF calibration.9 This is the share of EAD between each threshold line and the STS perimeter cut-off, reported in Table 2 as “% of Aggregated EAD”. The yellow curve shows the corresponding share of STS-relevant mortgage RWA (calculated as the area below the orange series) on the right-hand axis.

The effect is material. The estimated viable SRT perimeter is 6.5% of aggregated EAD at the current 10% STS floor, 17.9% at a 5% floor and 49.6% at a 2% floor, corresponding respectively to 25.5%, 52.3% and 84.2% of STS-relevant mortgage RWA. In amount terms, this is about €290 billion, €800 billion and €2.2 trillion. These figures are potential supply estimates, not issuance forecasts. Table 2 reports the full calibration ladder.

The evidence points to a clear calibration trade-off:

• The current 10% STS floor confines mortgage SRT to a narrow high-RW segment, representing only 6.5% of aggregated EAD.

• A 5% STS Resilient floor, as proposed by the Commission and Parliament, would broaden the market, but still leave more than 80% of mortgage EAD outside the viable perimeter.

• A 2% floor would be materially more effective, while still leaving roughly half of aggregated EAD outside the estimated viable SRT perimeter.

Figure 1. Distribution of aggregated residential mortgage EAD by RW and implied viable SRT perimeter under alternative senior RWF calibrations

Table 2. Potential SRT perimeter under alternative senior RWF calibrations

This also addresses the concern that a lower RWF could displace covered bonds. Even at a 2% floor, roughly half of aggregated mortgage EAD would remain outside the estimated SRT perimeter. A lower floor would therefore support complementarity, not substitution: higher-RW segments could support SRT, while lower-RW assets would remain natural collateral for covered bonds and other funding instruments. Nor would a lower floor remove prudential safeguards: STS Resilient treatment would remain limited to senior positions meeting structural resilience criteria, while tranche attachment, retained senior exposure and the securitisation formula would continue to constrain recognised capital relief.

Several limitations remain. The sample is limited to IRB banks with granular Pillar 3 disclosures. Pillar 3 data do not always separate EU from non-EU mortgage exposures, and reporting practices differ. The estimates also rely on portfolio-construction assumptions and exclude issuer-specific constraints such as leverage ratio, operational readiness and supervisory execution. A group-level cross-check against residential-mortgage covered-bond issuers indicates that most sample banks belong to groups with established mortgage covered-bond activity, reinforcing the complementarity point.

A lower floor without insurer access would create a market-access trap

Appropriate RWF calibration is necessary, but not sufficient. Mortgage SRT is long-dated and requires resilient long-term protection capacity. Funded investors are important, but most, except pension funds, are better suited to shorter-duration assets matching closed-end fund horizons. Residential mortgage portfolios can have maturities of 15 to 25 years, making them a natural asset class for (re)insurers with long-term risk-bearing capacity.10

This underwriting capacity is already relevant: by end-2024, (re)insurers provided UFCP on about 30% of the EU Non-STS market,11 with appropriate SRT recognition by supervisors.12 They do so through regulated claims-paying balance sheets, subject to Solvency II supervision, capital requirements and governance standards. In a banking-regulation context, the term “unfunded” refers to protection provided by regulated underwriting capital and diversified insurer balance sheets, rather than by posted cash collateral.

The companion SUERF Policy Note on capital cliffs explains why safeguards for insurer-provided protection should be robust at origination, rather than ongoing binary tripwires that create procyclical cliff effects. The narrower issue here is legal and economic coherence between SECR and CRR.

If STS Resilient mortgage transactions benefit from a lower senior RWF, but (re)insurer guarantees are excluded from STS Resilient treatment, banks will have a strong incentive to structure around eligible funded protection providers. This is not only because STS Non-Resilient treatment would deliver weaker capital relief. For synthetic SRT, STS criteria already overlap substantially with the proposed Resilient criteria, leaving the senior attachment-point test as the main practical differentiator. STS Non-Resilient may therefore be a residual and largely uneconomic category for mortgage SRT. A reform intended to broaden the European risk-bearing base could instead reduce the role of regulated insurance capacity in the economically usable STS market.

If SECR Article 26e(8)(aa) creates a safeguarded route for EU-authorised (re)insurers to provide UFCP in STS synthetic securitisations, CRR Article 243 should not neutralise that route by excluding those guarantees from Resilient treatment.

Three fixes are therefore essential:

• For STS Resilient treatment under CRR Article 243(3), any exclusion of guarantees provided under SECR Article 26e(8)(aa) should be deleted, so cross-references do not defeat the economic purpose of the reform.
• For Non-STS Resilient treatment under CRR Article 243(4), (re)insurer UFCP should remain eligible where providers are Solvency II-regulated, rated CQS 2 or better13 at origination, and diversified.14 All (re)insurers currently active in the Non-STS SRT market meet these standards; preserving this eligibility framework would maintain existing Non-STS protection capacity and the credit quality of current participants, while applying the same standard to future entrants.
• Eligibility safeguards should be assessed at origination, not through ongoing binary tests that could later remove STS or Resilient treatment even where the protection remains legally enforceable and prudentially managed. Later deterioration should be managed through ordinary credit risk mitigation, counterparty-risk capital, exposure limits, supervision and credible mitigation actions.

This is not a request to weaken prudential safeguards, but to make them robust entry conditions rather than market-closing tripwires. A lower RWF would make STS Resilient economically attractive for mortgage SRT; insurer access would provide the long-term protection capacity needed to scale.15

Policy conclusions

The EU Securitisation Package can create a genuine European mortgage SRT market only if STS Resilient is both economically usable and open to the protection capacity needed to support it. The 40-bank evidence shows that a 5% floor would broaden the current perimeter but still leave most mortgage EAD outside the viable perimeter, while a 2% floor would materially improve usability and preserve complementarity with covered bonds. A lower floor would not remove prudential safeguards: STS Resilient treatment would remain limited to senior positions meeting structural resilience criteria, while tranche attachment, retained senior exposure and the securitisation formula would continue to constrain recognised capital relief.

Calibration alone is not enough. If (re)insurers can access STS but not STS Resilient treatment, insurer-supported STS transactions could be left outside the economically usable mortgage SRT segment, even while SECR formally opens an STS route for them. The issue is therefore not only the floor level, but whether the capital framework allows the long-term protection capacity needed for mortgage SRT to participate in the viable segment.

A coherent reform should combine a usable 2% floor, clear (re)insurer access to STS Resilient treatment, continued UFCP access to Non-STS Resilient treatment, origination-date certainty and a sufficiently broad eligible provider base. In short, the Resilient label will only unlock SRT for mortgages, one of the largest and safest asset classes of EU banks, if calibration, legal coherence and access move together.