Following completion of its securities lending CDM pilot in November, ISLA plans to finish coding of its CDM minimum viable product for mid-June as part of a wider set of standardisation initiatives
The Common Domain Model — or CDM — provides a common data representation of transaction events, offering a common template or set of fields that the industry will use to share trade information and other key data.
This should reduce the burden of reconciliation and lower the risk of mismatches or settlement failure caused by inconsistency in how data fields are used.
For close to three years, the International Swaps and Derivatives Association (ISDA) has been active in the OTC derivatives market in promoting a common digital representation of the steps, or ‘lifecycle events’, associated with a derivatives transaction.
Subsequently, ISLA has been working with ISDA and the International Capital Markets Association (ICMA) to apply a CDM to securities lending transactions.
With this objective, it completed a pilot with REGnosys, its technology vendor for the CDM project, in November 2020. This securities lending CDM pilot concluded last year with a showcase event that highlighted a cross section of the benefits that can be realised through the adoption of a CDM for securities lending markets.
The pilot advanced at an aggressive pace, according to ISLA director of market infrastructure and technology David Shone, and this was possible, in part, as a result of the pre-existing foundation that ISLA was working with, developed initially through ISDA’s CDM programme.
The ISLA pilot project demonstrated the reusability of this model, meaning the ISDA CDM for derivatives transactions can be reapplied, with some refinement, to securities lending. “The design principles of re-usability and composability meant that we are able to construct product and event models quickly, re-using some components and augmenting them for our purposes,” says Shone.
As an active participant in the ISLA pilot study, FIS’ director of securities finance and collateral solutions Ted Allen says that the CDM pilot project delivered a “meaningful proof of concept in a compressed time frame” and was a fine example of how diverse players in the market can come together with a common purpose.
The primary aim of the pilot, notes Allen, was to model the allocation process as applied by an agent lender during a trade execution. To ensure the scope was achievable, this modelled a cash DvP trade from its inception and also focused on settlement actions and collateral provision. By the end of the pilot, the working group agreed on a set of core concepts for trade execution and allocations that are modelled into the CDM.
It was noteworthy, according to Allen, how often the pilot group identified aspects of the business that are not understood as well as they should be. “For example, block trades, where an execution has multiple allocations associated to it, came under scrutiny in relation to the actual terms of the agreement and practical implementation.”
ISLA’s Shone explains that adoption of an agile methodology for the CDM project created a fertile environment in which ISLA, REGnosys and subject-matter experts from the association’s membership were able to conceptualise, model and code quickly and effectively. “The goal was to create a model that could be used by member firms in the showcase event at the end of the pilot period,” he says. “This approach also provided flexibility to move faster in some stages than others, to re-prioritise as necessary while still enforcing scope boundaries.”
Key to the success of this pilot, Shone observes, was the ability to access data samples from ISLA’s membership, data that could be used for validating the CDM model. “Data transfer requests can be prolonged at times, so we ensured during our follow-on work this year that members of our working group were aware of data sample requirements in advance, thereby reducing delays in receiving those samples.”
According to Sharegain founder and CEO Boaz Yaari, the CDM pilot has shown how much appetite there is in the industry for greater collaboration, especially on the fundamentals. “It’s taken a huge effort from ISLA to get us to this point. But their determination and focus has paid off: with the CDM, we now have a clear path forward for the industry.”
With this framework in place, Yaari says that there’s no room in 2021 for any part of the capital markets to have diverging data structures. “We need a unified approach that can be applied across the entire securities lending ecosystem,” he says. “Look at PSD2 [the second Payments Services Directive] and open banking. As soon as you standardise those foundations of an industry, you pave the way for a wave of innovation. It’s already happened in payments and current accounts, now it’s happening in securities lending.”
Laura Allen, managing director at Trading Apps, reinforces these sentiments, stating that the CDM will provide a common language for market participants, ensuring they are using data fields in a consistent way and accelerating moves towards process automation.
This is important to improving connectivity between market participants, reducing the cost and complexity of communicating information across the SFT transaction value chain and strengthening opportunities to develop ‘plug and play’ solutions. “As a technology house that has pushed strongly for improved connectivity and automation across the transaction lifecycle, we look forward to the CDM being open sourced so that we can adopt it, contribute to it and provide support for the CDM framework within our products,” says Allen.
Minimum viable product
ISLA’s objective is to set in place the central elements of the CDM model, a ‘minimum viable product’ (MVP), for adoption by market participants by mid-2021. ISLA’s Shone says that this objective is actually close at hand. The MVP approach has built on the pilot, focusing on returns, re-allocations, non-cash collateral and billing. The working group itself has expanded significantly, enlisting further Tier 1 borrowers and lenders, but also one large institutional beneficial owner, one further market infrastructure firm and firms at the leading edge of fintech in the securities finance industry. “Development of the MVP and a chance to be part of it has resulted directly in new ISLA members,” says Shone.
Owing to the similarities between securities lending, derivatives and repo products, FIS’ Allen observes that it is becoming increasingly straightforward to describe contracts with a common terminology. The work being done by the Clause Library and Taxonomy Working Groups is building upon this foundation, actively defining a standard language that can be used across the market.
More broadly, their work is aligning legal documentation and contracts with the definitions in the CDM. Among other benefits, this establishes the conditions for users of the CDM to enter into electronic contracts. “Digitalisation of agreements like the Global Master Securities Lending Agreement (GMSLA) will enable document negotiation and execution to be performed without any human intervention, providing greater alignment between contracts and hence improving risk management,” says Ted Allen.
Widening CDM application
The evolution of the CDM will not stop at the advances outlined above. Inter alia, discussions are ongoing across trade associations about the prospect of modelling repo products and around standardisation of collateral schedules within the CDM.
What’s exciting about the CDM, says Sharegain’s Yaari, is that every part of the market will benefit from having a standardised base of activity. Each step in the securities lending lifecycle, from onboarding and contractual negotiations, through to collateral management and corporate actions, will experience benefits from that standardisation. Across the board, there is unnecessary complexity that the CDM can start to address.
According to ISLA’s Shone, this year has brought a renewed sense of collaboration between ISLA and other industry associations in furthering the goal of a truly common CDM covering multiple asset classes and markets. “The future should allow an equity swaps desk to hedge a short swap, selling short in the market, borrowing the stock to cover [the exposure], while treasury finances the whole thing through repos, managing interest rate risk through rate derivatives, all in the same code base!”
The common components, re-used across products, are likely to deliver major advantages to enterprise risk, finance and product control, as well as streamlining collateral optimisation by allowing an easily accessible common view across the entire firm.
“While clearly being some way away from this point, we will effectively have delivered the base securities lending functionality into the CDM within a year,” says Shone. “This rapid progress highlights how the CDM is now in a state where ongoing contributions can be applied quickly to multiple product systems and entities.”
As the model keeps improving, says FIS’ Allen, so do the benefits it can offer. From recent experience, the industry is aware how much effort was invested in the ongoing Securities Financing Transactions Regulation (SFTR) and Central Securities Depository Regulation (CSDR) initiatives and the cost impact has been significant in a time of tightening spreads and declining revenues.
“Had the CDM already been adopted, the effort and cost of SFTR and CSDR would have been significantly lower,” says Ted Allen. “This has been demonstrated by ISDA’s use of the CDM as a tool to help validate new regulations and we support the International Capital Markets Association’s (ICMA’s) efforts progressing CDM for repo.”
ISLA’s David Shone concludes that this strong foundation should result in the CDM becoming more attractive in the short term to any firm looking to re-develop their existing technology stacks. It should also appeal to firms wanting to make their service offering ‘plug and play’ across multiple counterparties by building to a standard that will be used by all.
ICMA is also developing the CDM to incorporate repurchase agreements (repo). “We continue to collaborate very closely on planning, working groups and common touch points,” says Shone. “Similar to our experience, we expect their model to benefit from some mutual uplift as ISLA’s work did from ISDA’s previous contributions.” All three associations are continuing to work towards a single common goal and towards an open-source model of contribution.
As noted, there are further potential benefits through standardisation of collateral schedules based on CDM. A major industry focus on ESG considerations, and the application of Uncleared Margin Rules (UMR) for derivatives transactions, have been important in driving attention to collateral optimisation and ensuring collateral meets expected ESG standards. ISLA, and its members have been working with ISDA’s collateral standardisation working groups on the CDM representation of a standard eligibility architecture.
This will improve the scope for optimisation of inventory and collateral across cleared, triparty and bilateral relationships for securities lending, repo and derivatives.
Concluding thoughts
Alongside these initiatives, Goldman Sachs recently announced the launch of a data management and governance platform, known as Legend, which is hosted in the public cloud on infrastructure maintained by the Fintech Open Source Foundation (Finos).
A number of other investment banks, including Deutsche Bank, Morgan Stanley and RBC Capital Markets, were involved in a six-month pilot to support data standardisation and collaborative data modelling, again building on the ISDA CDM model, using a shared version of the Legend platform. Now, with the Legend code available as open source, other organisations or individuals can operate their own implementations of the platform.
For FIS’ Allen, CDM will lower costs and help remove redundant processes such as reconciliations, break resolutions and fails management. The structures defined by CDM will also facilitate the use of blockchain in securities finance, given that distributed ledger technology (DLT) relies upon a backbone of consistent and standardised data and processes that are needed for smart contracts. “Without the common starting point provided by CDM, it will be more problematic for the industry to move to a DLT-based future,” he says.
Realistically, the marketplace may not see huge benefits from CDM for a year or two. Expectations for CDM are running high, but it will take time to model the complexities of the business and to gain adoption. “The three year headstart that we have due to the work already done by ISDA means that initial forays into the usage of the CDM can be undertaken internally, and here at FIS we are prioritising our approach for utilisation of the model for the future,” says Ted Allen. “It’s a long-term investment, and one that we are happy to make.”
Sharegain has an advantage that it started as an agent lender six years ago with a blank canvas, says the company founder and CEO Yaari. Consequently, it does not have legacy data structures to contend with. “The real value that I think the CDM will bring to securities lending is that we’ll be competing on the strength of our products, not on the underlying data or processes,” says Yaari.
“Securities lending is one of the last bastions of old tech in capital markets. The CDM is rolling down the drawbridge: it’s a foundation that will enable everyone to compete on the same playing field,” he concludes.
Digitisation pipeline
ISLA director of market infrastructure and technology David Shone updates SFT readers on ongoing work on the CDM MVP and the GMSLA Clause Library.
The MVP coding is due for completion in mid-June, with contribution of the code back into the full CDM shortly afterwards. This will make the ability to execute, allocate, reallocate, settle and return a trade publicly available to CDM users, whether the trade is cash or non-cash collateralised, and to bill for that.
The code contributed back to the full CDM will be three times the size of that contributed from the pilot in April 2021, not only including securities finance-specific components but also enhancements to existing code related to derivatives and some fundamental building blocks for future iterations of the CDM — for example, the ability to associate more than one legal agreement with a transaction (e.g. a Master Agreement such as a GMSLA, as well as an independent collateral agreement, say with a triparty custodian).
In tandem with the MVP work, we have been progressing our Clause Library initiative with our partners D2 Legal Technology. This has involved distilling the multiple variants of GMSLA language used by our members into discrete business outcomes and variables. Like the CDM work, our membership has made a major contribution to this with approximately 50 firms contributing to this effort. This initiative will conclude in August with the Clause Library being published for use shortly afterwards. It has been a pleasure to have not only one but two major initiatives on track to schedule successfully.
Ultimately, the Clause Library will need to undergo a digitisation effort that integrates it with the CDM, forming legal agreements that inform functional behaviour in the CDM, such as the outcome of a default event or a corporate action, on a particular transaction. There are multiple considerations regarding how to go about this and the right methodology to adopt. We will communicate further on this later in the year when that deliberation is complete.
