Sailesh Pandey (Chair)

Nomura Holdings

Tosha Ellison

FINOS

Rob Underwood

FINOS

Joey French

Intrinio

Anthony Golia

Red Hat

Steele Barcomb

Intrinio

Pascal Lefrancois

Innovestor

John Matera

FactSet

Mark Andrade

Fintech Sandbox

Joey French

Intinio

5 min

Convene & roll call

Introductions from the group included:

• Anthony Galia - Red Hat Chief Architect for FS data analytics and data in hybrid cloud

• Ben Sandler - AQR Security master development team

• Joey French - Intirnio co-founder, has API with lots of financial data and area of focus is proprietary securities and identifiers

• Steele Barcomb - Intrinio Dev director

• Pascal Francois - Innovestor, providing knowledge for enterprise industry classification and linking IDs. Acquired company in 2004 that had its own classifications, and would be useful to have open source GIX

• John Matera - FactSet Director of security master

• Mark Andred - FinTech Sandbox, representing 170 startups dealing with a number of data feeds; previously at Fidelity

5 min

FINOS, Data Technologies Program, and Security Reference WG Overview (Also Review Anti-Trust policy)

Introduced FINOS and the structure of programs, working groups and the PMC as well as standard working group practices and the anti-trust policy.

15 min

Use Case Walk Through 

The groups discussed potential use cases and areas where there are issues around ref data including:

• Aim to identify 10 to 12 use cases and then focus on a top 2 to 3 to solve

• Problems related to merging data, where sourcing from multiple vendors, e.g. non US equity options sometimes the vendors don’t have common identifiers like exchange tickers.

• Key challenge is the merging of data and being able to normalize identifiers enabling more efficient trade clearing. For FinTech Sandbox this is primarily equity and fixed income. One of the things they could, representing such a diverse set of groups, is poll their community for common challenges.

• Issuer data in EQ and FI is pretty fundamental and the data is even less standardized than in other areas

• One area of particular issue is security to entity relationship and mapping between non-prop identifiers. E.g. FIGI at core of security master and also grab data form SEC (central key entity identifier). This stil requires a lot of manual work to pull together. Look at ANNA (group related to ISIN standard) where they link ISINs to LEIs. While the the problem is general they tend to be more equity focused. 

• EQ is better served than FI but even in EQ there are inter-listed stocks, e.g. Vodafone has common shares with different codes for UK vs US and US has sponsored vs non-sponsored shares, which causes issues. 

• Reuters, FactSet and Bloomberg all have a commercial model aimed at solving some of these challenges. In the spirit of this forum is the idea to have open data as it’s expensive to maintain this? And how does this work outside of the commercial model.

• One extreme would be to have a golden record and and an OTC identifier that was open, e.g. generate an OSI code.

• It would be useful to understand what open data is out there. There is a tremendous amount of data out there and even linked to a security but many consumers of that data are trying to multi-source (and would still pay) but have to do extra work to link them.

• Another use case could be around combining data from various data providers and addressing the the challenge of linking them, other than using LEI, which has thin coverage. Even linking on CUSIP doesn't enable data sharing due to license restrictions so being able to share link that data would be key.

• Another challenge is mapping, e.g. and LEI or FIGI to a proprietary company identifier so the same securities map up to the correct parent company. 

• CUSIP is also commonly used but is extremely expensive and entities on both sides of a transaction have to be entitled. Also potential issues here where compliance uses CUSIP but a trader uses symbol so can cause errors so potentially not even standard usage within a company. Could CUSIP be open source so there is a single identifier that uniquely connects the securities. 

• Key problem emerging is merging securities data from different vendors and then across companies

• As you move outside US you also get to SEDOL. 

• Mark: US and Europe are also different and we do need it common across the globe. Yes.

• ISIN was also discussed and there can be issues with ISIN not being created quickly enough. It's also not unique across countries or even across exchanges. You could use MIC Code, CCY and ISIN to narrow down the specific listing, which means now you need more data. And for derivatives you don’t have ISIN until issue date. MiFID covers European listed trades and requires ISIN.

10 min

Top Issues Whip Around 

Discussion covered above.

15 min

Discuss issues and start to develop common prioritize and hypothesized set of first few issues to go after

The group agreed to document existing initiatives as well as identify specific use cases to help drill down to specific focus areas with broad interest.

5 min

AOB & adjourn