Key elements of the 2025 CCP Stress Test

Executive summary

The Bank of England’s (Bank’s) 2025 Stress Test of UK Central Counterparties (CCPs) is the Bank’s fourth such exercise. The Bank conducts regular stress tests of UK CCPs in order to assess their financial resilience, and to promote transparency and confidence in the UK clearing system. These exercises are exploratory rather than ‘pass-fail’ and the findings are used to identify potential areas of risk and to support and inform the Bank’s supervisory and regulatory activities.

The 2025 CCP Stress Test will focus on assessing the resilience of UK CCPs to the default of two or more of its members during a severe market stress. This includes a core Credit Stress Test and additional reverse stress testing and sensitivity testing that explores how the results change under increasingly severe assumptions. This year’s exercise will not include a full Liquidity Stress Test, but we will explore liquidity risks in more qualitative manner with firms and assess how risks have evolved since last tested. The Stress Test will also consider the impact on the wider financial system via initial margin and variation margin calls.

The Stress Test will be centred on a bespoke ‘Baseline Market Stress Scenario’, designed by the Bank. This scenario is an extreme but plausible hypothetical scenario, equivalent to a one in 3,500 event, in terms of where the scenario sits relative to the historic distribution. It is intended to represent extreme market moves over a period of two to five days in which there is escalating concern about fragmentation in global trade and sovereign debt risks. This results in a sharp decline in equity markets, rising interest rate expectations and mixed movements in commodities markets.

Alongside the Baseline Market Stress Scenario, the 2025 CCP Stress Test will include three additional ‘multiplier’ scenarios for the purpose of sensitivity and reverse stress testing. As we did last year, the Bank will also explore a wider range of hypothetical scenarios, including more extreme scenarios and those that break historic correlations. In contrast to the Baseline and multiplier scenarios, the Bank will use its own independent ‘desk-based’ modelling to undertake the revaluation of clearing member and client positions under these scenarios.

The 2025 CCP Stress Test will use a reference date of end of day Wednesday 26 March. Market price shocks will apply to market prices and member positions as of end of day on this date, and CCP resources (Initial Margin and Default Fund contributions) will be held fixed as of this date.

The Stress Test covers the clearing services of all three UK authorised CCPs (ICE Clear Europe Limited (ICEU), LCH Limited (LCH), and LME Clear Limited (LMEC). CCPs are required to submit the relevant data for the 2025 Stress Test to the Bank using data templates and instructions provided privately to them. The Bank will publish the results report in 2025 Q4.

Key elements of the 2025 CCP Stress Test

This document sets out the key elements of the Bank’s 2025 Stress Test of UK CCPs. This covers: (i) participation; (ii) the analytical components of the exercise; (iii) the market stress scenarios; (iv) data submission; and (v) disclosure.

This key element document is complemented by a published spreadsheet containing risk factor shocks for each market stress scenario. Participating CCPs have also been provided with ‘Instructions for completing the structured data templates’, ‘Structured data templates’, a ‘Supplementary data request’, and ‘BEEDs data submission instructions’.

Participation

All three UK authorised CCPs – ICEU, LCH and LMEC – will be in scope of the 2025 Stress Test, so it will cover six Default Funds and seven clearing services across three CCPs (see Table A). Temporary recognised non-UK CCPs and recognised non-UK CCPs are not in scope of the exercise.

Analytical components

The 2025 Stress Test will focus on credit risks to CCPs and the impact on the wider financial system via initial margin and variation margin calls. The core components of the Credit Stress Test will be complemented by further desk-based analysis aimed at testing a wider range of risks, building on the analysis of hypothetical scenarios conducted last year. In addition, the Stress Test will include a more qualitative exploration of how liquidity risks have changed since the last Liquidity Stress Test in 2023.

Credit Stress Test

The Credit Stress Test will assess the sufficiency of CCPs’ financial resources to absorb losses under a combination of the Bank’s market stress scenario(s) and the simultaneous default of selected clearing member groups.

As a starting point, the Credit Stress Test focuses on the default of the two clearing members whose default results in the greatest depletion of CCP resources (the regulatory ‘Cover-2’ standard). This will include an examination of the additional costs that CCPs would face when liquidating defaulters’ concentrated positions (‘concentration costs’).

The core Credit Stress Test will be accompanied by reverse stress testing, which applies increasingly severe assumptions to identify situations in which CCP resources could be exhausted, and sensitivity testing. This will include:

• Increasing the number of defaulters and examining a range of alternative defaulter (‘Cover-X’) combinations in order to assess whether resources sized against the Cover-2 standard are sufficient to cover the default of other combinations of Clearing Members.
• Increasing the severity of assumptions used to estimate concentration costs.
• Testing multipliers of the Baseline Market Stress Scenario – 1.5x, 2.0x and -1.0x – in order to assess resilience to much more severe shocks beyond historic precedents.

We will also extend the Credit Stress Test to consider a wider range of scenarios using a desktop modelling approach (see below for further detail on scenario generation). The objective of this is to assess a wider range of risks and identify the shocks that could pose the greatest threat to CCP resilience.

Member and Client impact

Alongside the Credit Stress Test, we will assess the impact on the wider financial system via initial margin and variation margin calls on CCPs’ members and their clients. This element was not included in the 2024 Stress Test, as these dynamics were separately being examined as part of the Bank’s system-wide exploratory scenario (SWES) exercise. The results of this element of the Stress Test will provide a more current view of potential liquidity demands in a stress and support the Bank’s wider surveillance and risk assessment.

Liquidity Stress Test

The 2025 exercise will not include a full Liquidity Stress Test but will explore how potential risks to CCP liquidity exercise may have evolved.

The 2023 Liquidity Stress Test found that all CCPs maintained a positive liquidity balance under stress, across all major currencies, even under very conservative assumptions about disruption to CCPs’ ability to mobilise liquid resources. Sensitivity testing explored the impact of wider disruption and identified that LME and ICEU were potentially more exposed to the failure of an investment agent whereas LCH was relatively more exposed if unable to liquidate non-cash collateral. These results generally reflect more structural and strategic decisions by CCPs – rather than the impact of a specific market stress – and so our exercise this year will probe whether CCPs’ approach has changed and whether we would expect to see substantive changes to the results.

Market stress scenarios

The Stress Test will be centred on a bespoke ‘Baseline Market Stress Scenario’, designed by the Bank and specifying shocks to around 900 market prices and rates. This Scenario is intended to stress different clearing services to a broadly consistent level of severity relative to observed historical market moves, while remaining plausible.

This year, we have revised our approach to scenario generation. Previously the scenario was grounded in specific historic episodes for each clearing service, scaled up to achieve the target level of severity in terms of aggregate profit and loss (PNL) at each clearing service. This approach had the benefit of preserving historic correlations between risk factors and achieving a consistent PNL impact, but the resulting market moves did not necessarily have the same level of plausibility across services; that is, some sets of shocks might be of lower probability as they were less representative of historic correlations.

In the new approach, we sample scenarios from a multivariate distribution fitted to historical market shocks. Given the number of variables involved, fitting this distribution is a multistage process:

• First, we reduce the dimensionality of the dataset using a principal component analysis (PCA) approach. PCA is able to divide the full set of risk factors into a smaller number of ‘components’ in a way that maintains most of the original variance.^{footnote [1]} In our case, we found that 80% of the variance of our 800 risk factors could be captured by only 40 principal components (PCs). These PCs are linear combinations of the input variables, so any linear relationships between the risk factors can be captured by a single PCA.
• Second, we fit a copula to the PCAs. While the components are linearly independent, the copula captures non-linear relationships between different risk factors, including tail dependence where the relationships between risk factors could change in a stress.^{footnote [2]}
• Then, we generate simulations of 100,000 potential scenarios from the distribution of principal components, that maintain the historical relationships. These simulations are then reconstructed into shocks to the original set of risk factors.

The resulting scenarios capture both linear and non-linear relationships between risk factors and match the historical distribution of the data. By generating a large number of scenarios and then selecting from them based on specified criteria, we obtain a final scenario that satisfies our criteria that the scenario should be extreme but plausible, consistent across clearing services and capture key risks. The figure below sets out the filtering process.

The final scenario that was selected represents a one in 3,500 probability event overall and is among the more severe of those identified in step 3 above. The final scenario sits beyond the 99.8^th percentile for each clearing service. The Bank has also carried out an assessment to ensure the scenario is broadly equivalent in severity across each clearing service.

As described above, the core Baseline Market Stress Scenario will be complemented by additional scenarios for the purpose of sensitivity testing, reverse stress testing, and exploratory analysis.

• We will ask CCPs to estimate the account-level impact of three ‘multiplier’ scenarios, constructed by applying linear multipliers to the individual risk factor shocks in the Baseline Market Stress Scenario. The first additional scenario represents an exploratory ‘opposite direction stress scenario’, in which all risk factor shocks are reversed.^{footnote [3]} This scenario is intended to allow for relatively simple exploratory analysis of a wider range of market stress shocks, without the requirements to build and extrapolate a bespoke market stress scenario. The second and third additional market stress scenarios are generated using 1.5x and 2.0x multipliers of the shocks in the Baseline Market Stress Scenario, and are used as part of our reverse stress testing. The prescribed two-day and five-day shocks for these three additional scenarios are also provided in the ‘2025 CCP market stress scenarios’ file.
• The Bank will also conduct exploratory ‘desk-based’ analysis to assess CCP resilience to a broader range of theoretical and exploratory market stress scenarios. These scenarios will be generated using the same methodology described above, but rather than specifying the direction of shocks we will sample more randomly from the tail of the distribution of generated scenarios in order to capture a wider range of risks. We will then use an in-house model to estimate the impact on CCPs and identify any vulnerabilities. This model is less accurate than the approach used for the core scenario – which is based on full re-pricing of positions by CCPs – but can be used to identify the types of risk that may threaten CCP resilience and support subsequent engagement with CCPs.

The 2025 CCP Stress Test will be run on a reference date of the end of day Wednesday 26 March 2025. This date is selected to be broadly representative of the level of CCP resources and exposures over the period since the Bank concluded its previous CCP Stress Test exercise. Market price shocks will apply to market prices and member positions as of end of day on this date, and CCP resources (Initial Margin and Default Fund contributions) will be held fixed as of this date.

Data submission

Data templates and instructions for both structured and unstructured data have been provided to participating CCPs. As part of the submissions process, CCPs are required to put in place processes to ensure high-quality data prior to submission to the Bank (including oversight by a person of appropriate seniority eg Chief Risk Officer). Once CCPs have provided final submissions, the Bank will run further validations and plausibility checks. Once the data quality checks are finalised, the Bank will run the stress test analysis. Data submitted as part of the 2025 CCP Stress Test may also be used for other supervisory and financial stability purposes.

Disclosure

The Bank intends to publish the results and findings from the stress-test exercise in a results report in 2025 Q4. As in the Bank’s previous exercises, this document will set out the approach taken in the exercise, analytical results, and the outcomes of the exercise. The Bank has a number of multilateral and bilateral co-operative arrangements with other regulatory authorities for oversight of the CCPs the Bank supervises, and (where appropriate) the Bank may also share relevant information on the key findings of the stress test under the provisions (relating to confidentiality) of such arrangements.