Seeing through DMIST

One of the many impacts of the early stages of the COVID pandemic was a massive sell-off in global markets over a just few trading sessions. The suddenness and violence of the sell-off exposed weaknesses in the operating models of the world’s largest exchanges, clearing houses, and brokers. As derivatives margin calls went through the roof, talk of defaults reverberated, amid insider chat that the clearing system was creaking more than authorities were letting on, and was possibly near breaking point.

The trauma of the March 2020 experience left many of the derivatives industry’s sharpest minds in little doubt that, while post-2008 regulation had done a lot to protect the system, it was still alarmingly fragile in the face of mega-spikes in volatility. With that in mind, efforts have continued over the past two years to introduce further safeguards, particularly into the post-trade environment.

One of the products of the 2020 experience was DMIST, or the Derivatives Market Institute for Standards, an FIA founded and industry-governed body focused on improving the efficiency and resilience of trade in listed and cleared derivatives. After a period of establishment, DMIST in November made its first major proposal to streamline processes in trading and clearing. The initiative focused on one of the more complex workflows in the post-trade space, which is where trades are allocated or given up. Successful execution of allocations depends on market participants executing a series of coordinated, sequential actions from end-to-end, which are reliant on operational processes and/or technologies to achieve integrated workflows. In practice, both coordination and workflows are subject to inefficiencies that are the natural result of market conventions that have evolved over time, rather being codified in rule books.

Recognising a pervasive lack of timely communication between clients and executing brokers, and executing brokers and clearing brokers, as well as final booking delays at clearing brokers, DMIST suggested a new standard called 30/30/30.

Under 30/30/30, all clients, executing brokers, and clearing brokers should submit and process their give-ups and allocation messages within 30 minutes of the relevant event that triggers their action. So, the client should provide allocation instructions to the executing broker within 30 minutes of a completed trade—the clock starting when the executing broker confirms the trade and stopping when the instructions are sent. Similarly, the executing broker should process client allocation instructions and should allege resulting give-ups within 30 minutes, the clock starting when allocation instructions are received from the client and stopping when the broker initiates give-up allocations to clearing brokers through clearing houses. And the same principle applies at the clearing broker, which has 30 minutes from the point that it accepts the allocation from the CPP to book the trade into the client end account.

The FIA believes that, properly executed, the new protocol should help the industry get the right trade into the right account at the right time, and remove many of the uncertainties and chaotic trade flows that exacerbated the March 2020 meltdown. The implicit question being asked in the consultation document is whether 30/30/30 is achievable, first globally and second when the market is in a highly febrile state.

Given allowances for a reasonable level of uncertainty, early indications of market reaction are tentatively positive. In an initial survey of DMIST participants, 71% agreed that 30 minutes is the correct time period for the initial phase of standardisation. Furthermore, 76% agreed that the time period should be the same for client, executing broker, and clearing broker. Still, at the recent FIA Expo in Chicago, 30/30/30 remained a topic of debate. While most people we heard from supported the rule in principle, many also expressed doubts. These centred not so much on the idea of communicating within 30 minutes but on the impact that a short communication window may have on data quality and reliability.

As many professionals working in the post-trade space will testify, the old adage of rubbish in rubbish out is as relevant as ever. Therefore, while in principle, running a clock on data exchange sounds liked a good thing, speed should not come at the cost of lower data quality. In short, if clients, brokers, and CCPs are rushing to complete processes in time, is there a realistic danger that data quality suffers, or that errors are missed? In raising the question at FIA Expo, market participants did not necessarily feel that data quality should be an absolute impediment to standardised timeframes. However, many were seeking comfort that the elephant in the room was not being ignored. As DMIST reviews the inputs to its consultation, no doubt it will consider whether there are indeed any trade-offs between 30/30/30 and suboptimal outcomes. All will be revealed in due course. The task for market participants, meanwhile, is to do what they can to increase reliability across the system. That should mean putting in place dedicated protocols and automated tools to ensure the risk of data errors is kept to a minimum.