Date

20 Aug 2019

Attendees

Agenda

1) Use Case reminder

2) Where we are on our road map.

3) Open Action Items

4) JIRA Issues Review - https://jira.edmcouncil.org/projects/DER/issues/DER-10?filter=allopenissues

5) Todays content discussion.

6) For next week.

Proceedings:

Discussion about whether this group should be concerned about the 3 use cases or focus back on development of the OWL ontologies. Pete suggested that we could flesh out the use cases enough to provide direction. John suggested that we split into two groups, one focused on use cases and the other focused on the content. The concern, though, is the slow pace given that we meet only every other week.

There is additional use case content that is not included in the posted skeletons, so we should definitely update them to include that much. Also, potentially we can go back to meeting weekly given that the EBRDF group seems to have taken a break. If that group is not going to resume, we could go back to weekly meetings after Labor Day. Even if EBRDF were to resume, perhaps they could use a different time slot. At the time, earlier this year, it made sense to work together, on an every other week basis, but perhaps we split that out. Focus on a specific use case at a time might help us deliver something in a 4-6 week time frame, although it likely would need to be much smaller than our current use cases.

Which use case should we focus on? Perhaps if we work towards a harmonized model for regulatory reporting, since that's smaller than risk management, and smaller than systemic risk, and likely to get lift more quickly. The question is really how we should sequence them. We would need to define what reports we should focus on, though we could create an extension to what was being done for EBRDF. If we can at least harmonize the terms that EBRDF started with, and also identify one or more specific reports that we can identify and show how we would address that, that could be a reasonable, smaller starting point.

In order for a regulatory reporting effort to have any real legs, we would need to engage one of the Swap Data Repositories, such as CME or DTCC or ICE to create some sort of prototype. That may not be doable in 4-6 weeks. For systemic risk, John could reach out to the Office of the Chief Economist, someone at the Fed, Treasury, etc. to get them engaged, and again, it would be a long term effort to implement.

From Jeff, and related to EBRDF, there is some progress being made on working on the content using ACTUS, so we could show how a portfolio of 100 swaps, including cash flows, could work. That might be related to a broader set of priorities - so do we work on that, or on other kinds of derivatives/asset classes, what should the priority be? Jeff's point is only that there was some work we could leverage as another perspective on the semantic content of the contracts, and other contract types would be interesting from an ACTUS perspective.

A common model for regulatory reporting would be more 'complete-able' in a shorter time frame, as it would not require us to expand into a broader set of risk ontologies, which we have not really done to date. To do counterparty exposure and systemic risk - would open up other opportunities, beyond derivatives. Independently of FIBO, there is an OMG working group for risk, which we could also leverage. It may be a simpler exercise to look at risk from an operational perspective. Even with respect to regulatory reporting, we should start with a specific target. Regulatory compliance is on every bank's short list, which may mean that it is the most prudent and manageable use case to focus on.

If there is work that has been doing at Treasury or the Fed, perhaps we could leverage that and get their participation to address something doable in a reasonable time frame (John). We aren't necessarily looking for regulatory change, for example for CFTC to make a change to require use of the ontology, but to show an example of what's possible. If we can get an understanding of how much involvement Treasury or the Fed has, then perhaps that's not the best approach.

There are at least two points of view on regulatory reporting - (1) regulators harmonizing the content they receive from reporting parties, which is difficult to do and which FIBO could help with, (2) ensuring that the reporting parties use the harmonized vocabulary for reporting from the outset, which would assist the regulators and even the banks in understanding the content that they are receiving, (3) more generally, for organizations to bring together the content required for specific regulatory reports internally, such that the content is semantically enhanced / harmonized earlier in the food chain. Perhaps we could work with an SDR to do a pilot that would provide RDF to the CFTC rather than provide their reports in the current hodge podge way they do today. If CME would engage with to try to prototype something, and then propose to the CFTC that they deliver content to them in RDF - that might have some potential.

Work on counterparty exposure could broaden the set of contract types we look at, although looking at something like notional balances between the parties might limit the size of the effort, rather than use of RDF as the payload for messaging transactions (Jeff). Transaction reporting is already in work via CPMI-IOSCO, via ISO 20022, etc., although perhaps they don't include the semantics we would be interested in today. FIBO could be a better tool for looking at relationships between counterparties and contracts. There is a weekly call on Fridays at 6 am looking at harmonizing certain aspects of reporting across a number of existing standards - Richard Beatch is on that call.

It's probably a sequence of steps that we need to execute on. Once we've defined the contract semantics we could add the counterparty exposure extensions, and look back at the other ontologies, such as in BE, to improve on those.

Decisions:

Action items

Browser not supported