Follow us

Tackling the next phase of Initial Margin

Over the next two years, the regulatory framework governing margin and collateral will continue its march toward full implementation, and the financial industry will confront new challenges as it prepares for the operational and compliance changes that result from the next wave of initial margin (IM) requirements.

A group of industry experts gathered for the 2018 Collateral Management Forum, which was hosted by DTCC-Euroclear Global Collateral Ltd. (GlobalCollateral), to discuss how firms will manage larger margin call volumes, tri-party versus third party account structures for segregating collateral and the effect on the speed of settlement.

The panel discussion, entitled, Industry Leaders Tackle the Challenges: Streamlining Initial Margin, was moderated by Amy Caruso, Chief Commercial Officer, GlobalCollateral and featured prominent industry experts.

Caruso asked the panel to elaborate on how the industry is anticipating the effect of the margin rules that require an account that never had to post before, to now, do just that.

The main takeaways from the panel were:

  • Firms need to understand the numbers that will help them identify which parties and counterparties will have to post IM.
  • Firms may need to have the proper documentation at the ready if the counterparty is in scope, regardless of whether or not it is expected to actually reach the $50 million IM threshold.
  • Collateral settlement must move from a low-tech, manual environment to a more efficient, automated system such as GlobalCollateral’s Margin Transit Utility (MTU).

Aggregate average notional amount (AANA) and IM Thresholds

As the IM rules get phased in, several numbers are critical to their proper application: the AANA or average aggregate notional amount; the IM threshold and the minimum transfer amount.

Beginning in 2016 and concluding in 2020, the mandatory IM requirements are being phased in based on legal entities’ AANA. Phase 3 went into effect September 1, 2018 for covered swap entities combined with all its affiliates. For Phase 4 in 2019 the AANA is $750 billion. For Phase 5 in 2020, the AANA is $8 billion.

With each phase, a larger number of counterparties are expected to come into scope. Identifying those clients that exceed both the AANA and the $50 million threshold per counterparty pair is a major concern. As one panelist stated, those are the clients that are going to have to exchange margin and have margin posted from their dealer counterparty.

Another challenge in determining who has to post IM is trying to decide how close the parties are to hitting the $50 million threshold amount without the use of a well-defined calculation method. One panelist noted his opinion for the need to have a document in place if the counterparty is in scope, regardless of whether or not you expect them to hit the $50 million.

Tri-party versus Third-party accounts and STP

The rules, created in an effort to mitigate risk, also call for segregation of IM. There are two types of structures to accomplish that goal: the tri-party construct and third-party accounts.

Third-party accounts are a better fit for the buy-side, according to one of the panelists. And according to another panelist, the tri-party model is more efficient and cost-effective from the sell-side standpoint. For those legal entities that choose the third party account structure, they will need to have an efficient method to release collateral, and the current process of using faxes is not sustainable, as discussed by the panel. The MTU is based on a market practice that notifies the relevant custodian that both parties agree to the release and therefore, the instruction does not require a follow-up fax/authorization process.

The quicker turnaround time for settlement, especially in light of the T+1 settlement process, demands the use of automation to increase efficiency. The entire panel agreed that the industry has to replace faxes with high-tech alternatives such as GlobalCollateral’s MTU, which allows straight-through processing.