This is the second in our series of articles generated from a recent Calypso webinar on BCBS-IOSCO regulatory initial margin (IM) requirements.
Given that one of the first decisions to be made is how to calculate IM and exchange it with counterparties, we look at how you can benefit by using a standard methodology in the shape of Isda’s Standard Initial Margin Model (SIMM).
SIMM, which was developed in conjunction with Isda members, has now gained both regulatory and market acceptance in jurisdictions around the world, and is expected to gain further traction as larger numbers of firms come in-scope over the next two years.
In the course of our webinar, Isda’s Rahul Advani, director of public policy, and Nnamdi Okaeme, director of risk and capital summarised the SIMM approach and advantages it offers firms coming in-scope.
SIMM uses a sensitivity-based approach, with market risk as the primary input, rather than trade contract details. This means it can accommodate all trade types, no matter how complex - provided their sensitivities can be determined.
The SIMM approach is a parametric value-at-risk (VaR), predicting the likely change in value of a portfolio over a ten business day period, specified by the BCBS as the margin period of risk. This is basically the time it would take you to close out exposure and put in place a cushion to protect against possible counterparty default.
It is a prediction-based approach that needs to be set with a 99 percent level of confidence – again, in line with BCBS standards.
The advantages of using a single, standard methodology like SIMM will become even more apparent as more and more firms fall in-scope in 2019 and 2020, as if offers:
- Operational simplicity
SIMM offers a single methodology, including common weights and correlations, as opposed to the possible multiple models that would result if every firm did its own thing. This is particularly good news for smaller firms coming in-scope in Phase 4 and Phase 5, who may have limited resources.
- Reduced potential for disputes
IM requires you to calculate two numbers:
a. What you think you owe your counterparty
b. What you think your counterparty owes you
Your counterparty is also calculating these numbers – and if their numbers don’t match yours, a dispute will result.
A standardised model like SIMM can help to reduce the number of disputes, as both sides will be able to replicate disputed calculations and reconcile the differences on a common platform.
- Ability to handle additional regulatory margin requirements
The IM calculator is a fundamental building block in the compliance process and as such, needs to be implemented well in advance of industry testing. A simple margin calculator is not up to the task as regulatory compliance margins must also be calculated, including regulatory inputs such as multipliers and add-ons.
SIMM is a standard, parametric VaR model, but it also includes an additional margin component, based on regulatory margin add-ons and multipliers.
For example, regulators may consider certain product classes particularly risky at a given time and in given market conditions and so may ask firms for extra IM for these specific product classes (multipliers). Alternatively, they may ask for ‘add-ons’ across all product classes. Or they can include both additional components.
Additional IM may also be required bilaterally. Regulators may believe your counterparty carries more credit risk than can be observed, so requests more IM as a cushion against default.
- Consistent regulatory and governance oversight
This is an important aspect of BCBS/IOSCO guidelines.
The SIMM Governance Framework provides three key processes by which it will be reviewed, amended and developed in conjunction with SIMM users and regulators.
• SIMM quarterly margin monitoring
Firms perform tests each quarter (the Isda ‘1+3’ back-test, for example) to assess the performance of SIMM for their portfolios. The resulting data is submitted to Isda for analysis and is then anonymised and shared with participants and regulators to see how well SIMM is performing.
If the data reveals problems in the way SIMM is performing, and these are determined to be ‘persistent, material and widespread’, then action will have to be taken and changes to SIMM may have to be made – new risk factors introduced for example or parameters updated.
Following any changes, a recalibration exercise will be required.
• Annual benchmarking and back-testing
Benchmarking involves looking at observable standards for IM calculation and comparing them with SIMM. You need to do this for products that are cleared but that are also traded OTC so you can determine the IM from the CCP and compare it with the SIMM IM.
An advantage of performing and feeding into these processes is that you can leverage the results for your own internal governance purposes – for model validation or when preparing for meetings with regulators, for example.
As well as implementing SIMM (or another model), firms will also need to go through an approval process with local regulators. Back-testing and benchmarking are part of this validation process.
The aim is to check that the SIMM you calculate at the end of a quarter is conservative enough. Isda has adopted the ‘1+3’ approach for back-testing.
This means incorporating a one year stress period that is generally determined from the calibration of the SIMM. A three-year period is also calculated; the most recent period from a given portfolio reference date.
You then determine the relevant P&L vectors for the same’1 + 3’ period and compare these against SIMM. SIMM should be equal to or greater than all the P&L vectors over this historical period, with 99 percent coverage.
The Isda back-testing approach produces two outputs:
• A colour, in accordance with the Basel traffic light test: red-amber-green. This indicates the way SIMM is performing, where red is not good.
• A quantification of the poor performance in the case of red or amber, to produce the shortfall amount.
Being able to benefit from a market standard methodology such as SIMM is a big help in the compliance process. But the IM generated will only be correct if the inputs to the calculation are correct.
The inputs required for SIMM calculations are sensitivities computed across a number of risk factors. Calculations must be performed in accordance with SIMM best-practice so risk inputs are generated consistently - otherwise you could become embroiled in managing endless disputes.
The size of this challenge should not be underestimated.
SIMM risk factors are used for margin calculations, but they also need to be accessed for disputes and validation. Calculations on the fly and recalculations will also be required, as will the ability to drill down at trade level or risk factor level.
You will also need to be able to produce results in the Isda Common Risk Interchange Format (CRIF) to facilitate dispute management and reconciliation.
Don’t leave testing till the last minute
An early start for testing is advised – not least because the early stages of testing can involve lengthy discussions about how to calculate Greeks: everything from the curve generation to how the pillars are defined and the cross-currency basis risk is calculated.
Industry testing is particularly important. At each stage, you will need to ‘battle-test’ your solution by selecting a number of friendly counterparties and examining in detail how you would reconcile risk factors for complex portfolios.
The good news is that Calypso’s professional services team has worked with firms coming in-scope in Phases 1, 2 and 3 and this experience means we offer particularly valuable help in testing to support you in your compliance journey.
This leads us into the third article in our IM series, which summarises exactly how and where our Professional Services team can help you– particularly when it comes to gaining regulatory sign-off.
Access the full on-demand webinar here.
Contact us to discover how we can help you achieve IM compliance and optimise collateral management.