Enhancing Initial Margin Efficiency With Innovative Risk Models

In an environment of increasingly stricter regulations and risk controls, and a general upward pressure on the costs of members to clear, higher-margin efficiencies are as important as ever. Modern initial margin models have the potential to lower the members’ funding costs while at the same time confidently provide more precise margin calculations.

In the last couple of years, Filtered Historical Simulation (FHS) and Monte Carlo (MC) based initial margin algorithms have increasingly gained traction, where several of the largest CCPs have moved away from legacy methodologies like SPAN. Leaving the ones who have not yet taken the step in risk of falling behind by losing competitiveness.

In FHS models, the beauty lies in its simplistic nature where the basis of the instrument scenarios can be directly tracked to historical market data, and where individual correlations are inherent so that the resulting portfolio margining is a simple aggregation of its position's profit/loss values. MC on the other hand tends to be a bit more complex, although having the advantage of being able to simulate close-outs in a fashion where also connected collateral portfolios are included.

Moving away from legacy margining models, the parameters to estimate and reference data to maintain decreases substantially. This simplifies the day-to-day work on administration and governance around parameters and configurations, leaving the CCP’s risk division to spend more time on exploratory and innovative work than routine.

FHS and MC models open up a great potential to visualize and present the components behind the margin numbers that can give an easy intuitive understanding. Coupled with easy drill downs into the calculation steps, it can save CCPs valuable resources otherwise used to respond to member enquiries on why their margin ended up as they did. Instead of lengthy email conversations and exchanging Excel sheets, drilldown functionalities allow members to dive into the calculations all the way from the scenarios, through profit/loss values of its portfolio and to applied add-ons - to easily understand exactly where its total margin came from.

Combining tailored dashboards and easy-to-navigate risk tree structures for a bird's-eye view, with configurable alerts and notification functionality, the users get a unique position to effortlessly monitor risk. For both internal risk managers and member users, having powerful tools to quickly identify increased risk while at the same time be able to drill down into the drivers of the aggregated risk is invaluable.

Inevitably, more sophisticated models put high requirements on computational resources. With a scalable microservice architecture using the most modern technology stack, initial margin calculations can still be done in an instant. This provides a CCP solution that effectively calculates risk in real-time, across initial- and variation margin as well as collateral values. The CCPs can then immediately identify deficits and margin breaches, to quickly start the work to ensure sufficient funds are in place to maintain the exposure.

In recent months, regulations aimed at increasing margin transparency have been proposed by the Basel Committee on Banking Supervision (BCBS), the Committee on Payments and Market Infrastructures (CPMI) and the International Organization of Securities Commissions (IOSCO). The
suggested regulations address the challenges with more uncertain and volatile initial margin requirements, where its procyclical nature in especially FHS models can lead to sharp increases in initial margins, and often at times when liquidity is tighter. A key need identified were more accessible margin simulations tools that have hypothetical/what-if capabilities. If the proposed regulation takes into effect, many CCPs would need to provide initial margin data more accessible and expand their margin simulation tools.

With modern models, the discrepancy between the purpose of initial margin to estimate the collateral to cover future extreme but plausible market conditions, compared to its lagging responsiveness to market conditions, have widened. With its procyclicality nature, modern initial margin models provide lower initial margin in calm and normal market conditions, then quickly, although retroactively, responding to the shifts in market volatility. This delay means that changes in market volatility take effect in the initial margin calculations the following day.

What really would benefit clearing members and their clients are systems capable of providing quicker indications of initial margin increases to provide indications and forecasts, factoring in increases in market volatility, on how the initial margin will change. With efficient systems handling high computational loads much more efferently, the time between the change in the input parameters and the resulting initial margin can be substantially shortened.

Moving to a modern margining methodology unlocks a broad range of benefits both for the CCPs, its members and their clients. This move is particularly advantageous when partnering with a vendor that has business expertise and technical capabilities. Such a partnership enables the provision of state-of-the art clearing and risk platforms, that can efficiently handle the calculations as well as providing an unparalleled user experience with powerful dashboards and visualizations.