ECC Prisma for Commodities
Introduction
The rapidly changing geopolitical climate and recent volatile years call for more stability and integrity on the commodity markets. To meet these challenges, we are committed to future-proofing our risk management capabilities, while ensuring capital efficiency.
As part of our efforts to continue to serve as a resilient and forward-thinking clearing house, we are now transitioning to a new, more sophisticated portfolio margining model. “ECC Prisma for Commodities” is a cutting-edge Value-at-Risk (VaR) based derivatives margin model, which will leverage advanced portfolio-based risk assessment to provide a more accurate and holistic view of risk.
It has been designed to better recognise the true risk of a portfolio by accounting for the offsetting effects of different positions. For many clients, this will translate into more accurate margining and significant potential capital efficiencies through superior netting capabilities.
The core of ECC Prisma for Commodities is a risk factor-based scenario price generation, which enables the construction of correlated risk factors to generate product and portfolio P&L distributions. Products sharing the same risk factors and which will be liquidated jointly, are grouped into liquidation group splits. Liquidation group splits are an important feature of the model, as diversification benefits are only granted within these groups and not between them.
Initial margin is a forward-looking margin component that quantifies a provision for potential future losses over the liquidation period of a defaulting member’s positions. The calculation of initial margin assesses the risk for each margin account’s portfolio of derivatives contracts, considering potential hedging effects for positions within the portfolio.
Technology
The new model is based on Prisma technology, originally developed by Eurex Clearing AG, and will run on Deutsche Börse Group’s Risk Management Platform R7 which is the Next-generation Risk Management Platform fostering technology leadership. The risk infrastructure, technology and data management of R7 enable scalable, consistent and fit-for-purpose risk management services, strengthening its ability to offer dependable and stable CCP services to the market with a variety of products and client types. The inherent design is to cover a broad range of risk calculations for CCPs and improve operational efficiency for cloud-based operations. Based on the Prisma methodology, the margin simulation tool is complemented by comprehensive margin replication and initial margin calculation functions.
This move is more than a technical upgrade. The new, more flexible model will provide a platform upon which we will be able to expand our services to serve changing clients’ needs and ensure that all our customers are well-positioned to benefit from future market opportunities.
Throughout the implementation process, we will provide regular updates on upcoming milestones. The new methodology has already been validated by the regulator and technical implementation is ongoing.
Key Benefits
Methodology Overview
In the future, ECC Prisma for Commodities calculates initial margin requirements for all derivative contracts cleared by European Commodities Clearing AG. Cleared products sharing similar risk characteristics and aligned default management processes are grouped together into one portfolio. Comprehensive portfolio risk calculations consider hedging effects and, thus, lead to more efficiencies in margin requirements and potentially associated capital costs.
Liquidation Groups and Liquidation Group Splits
The Initial Margin for derivatives is calculated separately for each margin account's portfolio of derivatives contracts. Since a derivative portfolio for a margin account might contain products that cannot be liquidated together, it may need to be further divided into disjoint sub-portfolios (Liquidation Groups) to ensure that offsets granted in margining can be materialized during the liquidation of a portfolio in case of default. These Liquidation Groups are further broken down into Liquidations Group Splits, which are the most granular level on which the Market Risk Initial Margin component is calculated and, thus, where portfolio margining effects can materialize.
| Hierarchy of Initial Margin Calculatoin and Aggregation | ||
| Level | Description | Risk Measure Component |
| Collateral Pool | Margin Requirements are collateralized at Collateral Pool level. Clearing Members can assign different margin accounts to the same Collateral Pool, benefiting from net margin calls at the lower asset / collateral segregation possibilities. | Margin Call (aggregated Margin Requirements are covered by one collateral value, only net shortfalls are called) |
| Margin Account | Clearing Member portfolios are further split into proprietary and one or many General Omnibus Segregated (GOS) / Simple Omnibus Segregated (SOS) client accounts or are even individually segregated (ISA) for Non-Clearing Members (NCM). This setup allows for an autonomous handling of such margin accounts during a potential liquidation event. | Total Margin Requirement (prior to aggregation, margin credits are floored to zero on margin account level (gross margining), allowing for proper client asset segregation and potential porting) |
| Liquidation Group (LG) | Liquidation Groups contain only products which can reasonably be liquidated together because they share common risk characteristics and can be priced by a suitable number of auction participants. They follow the same liquidation process as defined by the default management process under consideration of the applicable segregation and portability arrangements. | Initial Margin Requirement (aggregation of all individual margin components from LGS levels) |
| Liquidation Group Split (LGS) | Liquidation Groups are further subdivided into Liquidation Group Splits (LGS) to ensure that portfolio margin offsets are granted only between products that show a significantly stable statistical correlation or an economic rationale for correlation. Additionally, a separation into liquidation group splits might be needed to account for different model parameterizations (e.g. holding periods, confidence levels, risk measures, etc.). | MRIM (max of FHS and SP VaR incl. CMAs) LA PMA |
Initial Margin Components
The forward-looking margin is called initial margin (IM). It estimates the potential future exposure of clearing members' portfolios held in the respective margin accounts in alignment with the default management assumptions. More precisely, it employs the decomposition into liquidation groups and even more granularly into splits which reflect the respective margin period of risks and the pre-defined confidence levels.
Initial margin consists of the following main components:
| Model Components of ECC Prisma for Commodities | |||
|---|---|---|---|
| MRIM (Market Risk Initial Margin) | LA (Liquidity Adjustment) | PMA (Portfolio Margining Adjustment) | |
| FHS - Filtered Historical Simulation | Value-at-Risk | ||
| Compression Model Adjustment | |||
| SP - Stress Period Simulation | Value-at-Risk | ||
| Compression Model Adjustment | |||
Market Risk Component (MRIM)
The core market risk initial margin (MRIM) is based on two pillars. A “pure” filtered historical simulation approach and a Stress Period VaR capturing periods with extreme market conditions that can be outside of the current lookback period. The combination of volatility filtering and the consideration of stress periods ensures that margins are not decreasing to unacceptable levels and thus, reduce procyclicality, without losing the necessary reactivity in extreme situations. A dedicated Compression Model Adjustment (CMA) compensates for inaccuracies and model errors, for example when modelling implied volatility surface scenarios for options.
The market risk component targets a confidence level of 99% for listed products. The pure market risk components, i.e., without model adjustments, are calculated based on 750 filtered-historical scenarios with a three-year lookback period and 250 stress period scenarios.
The core calculation steps for the Filtered Historical Simulation are as follows:
| Step | Result | Description |
|---|---|---|
| Modelling of historical risk factors | Risk Factors | For all relevant risk factors (e.g. future prices, implied volatalities, FX rates, interest rates) historical return time series based on market data are calculated. |
| Filtering of historical risk factor returns | Filtered Returns | Return time series are normalized to ensure that the derived filtered hostorical return data reflects the current volatility regime. |
| Scenario risk factor values and scenario prices for individual contracts | Scenario Prices | For each contract, a theoretical scenario price is calculated for each historic scenario based on the attributable shifted risk factors. |
| Scenario P&L Calculation | Scenario P&L | For each historic scanrio date and each contract, the scenario P&L is given as the price difference between the scenario price of the contract and its current market value. A weighting factor ensures that contract specific risk profiles are adequately considered. |
| Sample Generation | Samples | Samples are constructed by aggregating over all contract-level scenario P&Ls multplied by the relevant position effects size in the portfolio. Hereby portfolio effects materialize as the P&Ls are consistently (scenario-by-scenario) aggregated. |
| Calculation of the risk measure (FHS VaR) | FHS Risk Measure | The final risk measure, in this case the Value-at-Risk, is read off at a defined confidence level. |
While the filtered-historical component ensures model reactivity in light of changing overall market volatility, the stress period component is included to ensure margin stability and to counter procyclicality.
The tail risk measure Value-at-Risk is applied to the profit and loss distributions of the historical and the stress period sample individually.
The Market Risk Initial Margin component is finally determined by maximum of the two pure market risk components adjusted by Compression Model Adjustments (CMA) accounting for the data compression techniques applied within risk factor modelling.
In addition to MRIM, additional model adjustments like Liquidity Adjustments (LA) and Portfolio-Margining Adjustments (PMA) are introduced to further enhance the risk-adequacy and stability of the model. They explicitly address potential implications of model assumptions (PMA) or types of risk beyond market risk (LA).
Liquidation Adjustment (LA)
The Liquidity Adjustment (LA) accounts for effects when liquidating potentially concentrated portfolios and is designed to consider additional potential risk that may materialize during the liquidation. The model adjustment is characterized by the following:
- Dependency on the current level of market risk in the respective sub-portfolio. Therefore, the liquidity risk adjustment builds upon and scales the market risk component of the initial margin, i.e., the higher the market risk of an instrument’s price, the higher the premium.
- Dependency on the relative size of the sub-portfolio. The Liquidity Adjustment is a function of the sub-portfolio size and the total market capacity in the respective market.
- Market capacities and associated model parameters are market-specific and reflect the risk and trading characteristics of the instruments contained in the respective market.
Portfolio-Margining Adjustment (PMA)
The Portfolio-Margining Adjustment enables enhanced steering of offsets within the Initial Margin calculation on LGS levels. The PMA shall ensure that the offset between products with different underlying is below the allowed portfolio-margining reduction cap.
Latest News
| Publishing date | Title | File |
|---|---|---|
| 2026-01-12 | ECC Clearing Circular 02/2026 | ECC Prisma for Commodities: Introduction of new derivatives margining model | pdf (229 KB) |
| 2026-01-12 | ECC Press Release - ECC advances transition to portfolio-based initial margin model | pdf (152 KB) |
| 2024-02-01 | ECC Clearing Circular 11/2024 | Announcement: Initial margin model transition to a portfolio-based approach | pdf (110 KB) |
| 2024-02-01 | ECC Press Release – ECC transitions to portfolio-based initial margin model to better reflect risk and further improve transparency | pdf (131 KB) |