The International Swaps and Derivatives Association (Isda) says a platform it developed with law firm Linklaters to automate the negotiation of derivatives documents can help firms move towards adoption of smart contracts for over-the-counter derivatives trades.

The companies launched Isda Create last year to allow market participants to modify standard documentation and distribute it to counterparties. The platform enables the capture of structured legal data and was developed by Linklaters’ internal technology startup, Nakhoda.

“This is more of a precursor to smart contracts,” says Andrew Kayiira, director of commercial product development at Isda. “Before you get to smart contracts, you have to have consistent, reliable data, allowing you to systemically define how the contracts will work.”

Isda Create is a web-based platform that automates the negotiation process and provides a downloadable audit trail that enables users to view any changes that have been made. Though the platform exhibits some of the fundamental characteristics of a smart contract, Kayiira says true smart contracts imply the automatic execution and subsequent processing of terms and obligations between parties. However, he adds that Isda Create lays the foundations for smart contracts by enabling market participants to negotiate contracts online while retaining the flexibility of contracting on paper. In addition to greater efficiency, firms will have access to the data representations of those underlying contractual terms.

Kayiira says Create is intended to expedite the negotiation process, provide transparency, and enhance communication, but it does not automatically execute contractual terms or trigger the settlement of those specified obligations. Firms still control the legal terms they wish to agree upon, and can add their own provisions.

“What does make this a precursor to smart contracts is the further standardization of legal provisions, and the decomposition of those terms into structured data representations that assist in the operationalization of legal data,” he says.

Isda has published guidelines on smart derivatives contracts for technology developers, lawyers, and other stakeholders, and its work with Digital Asset has also had implications for derivatives smart contracts.  

Luke Martin-Fuller, product manager at Linklaters Nakhoda, says smart contracts could make decisions in reaction to real-life events years after being signed.

“To fulfill their potential, you would need to design and develop them in a way that is consistent across applicable legal, technological, and regulatory standards, and [that] works for all participants,” Martin-Fuller says. “[Isda Create] is the first step in that direction. We are bridging the gap between the paper world, which is where people feel comfortable, and the world of legal data. In a sense, capturing accurate and structured information about documents and their attributes is the first step on the road to smart contracts.”

Traditionally, the negotiation process involves numerous emails and phone calls, none of which are easy to track, says Hannah Patterson, managing associate at Linklaters. Then, after the document is finally agreed, data needs to be extracted and input into internal management systems before it goes live—a process that Patterson says can take upwards of 24 hours.

Isda Create automates the downstream flow of data directly into a user’s internal management systems, which also reduces the risk of human error.

The platform was initially set up to help users meet regulatory requirements for initial margin arrangements. However, in June, Isda announced new modules for the platform that allow generic amendments, and the addition of investment book of record (Ibor)-related documents and Isda Master Agreements.

The Create platform will also incorporate Isda’s recently-launched Clause Library, which sets out standard drafting options for frequently-negotiated provisions in the Isda Master Agreement, as well as their most common variants.

Capturing Data                                                                     

Every document on Isda Create has an associated data model. Data is captured in a hierarchical format that reflects the content of the document and any optionality within it. At the end of a negotiation, users can transfer the data from their agreements into their internal systems, providing a standardized format for the data as it comes out of the platform.

Currently, Kayiira says, firms extract attributes manually from PDF forms. So, for example, if a firm has 100 legal data points that are important, it often only captures 30 or 40 of them, because doing it manually forces them to rationalize how much of their legal contract they want to be accessible. With Isda Create, the user gets a complete view of the document as data, which can be leveraged by trading, risk, operations, and compliance teams.

The Nakhoda team works with Isda and Linklaters’ derivatives specialists to identify all the commercial and legal data points that must be agreed upon to execute an Isda document. The next step is to take each of those data points, group them together and represent them as digital “elections,” which are the choices users need to make under the derivatives documents in order to agree to the document.

Each election has implications for the drafting and, ultimately, the meaning of the document. The system breaks the contract down into several elections that, when completed, result in the user’s version of the draft contract.

Bringing in the Code

Another Isda initiative that complements Isda Create is the Common Domain Model (CDM), which is a blueprint for how certain common processes and tasks are performed in the derivatives markets. The CDM has seen patchy take-up among banks and market infrastructures since it was conceived in 2017, as some banks are said to be having trouble making a case internally for a project that promises savings of up to $3 billion a year on post-trade processes, but generates no revenue.

At its core, though, the CDM aims to provide the executable code components and definitions for clauses and calculations found in Isda’s derivatives documentation and other market and lifecycle processes, says Ian Sloyan, director of market infrastructure and technology at Isda.

“Those components can be put together when one executes a smart contract,” Sloyan says. 

Isda is currently working on methods of integrating CDM output with Isda Create, but the conversion of data from Create into a CDM format is possible now. 

“We have already built that in the CDM and have demonstrated it works, and the examples are in the CDM open-source platform. We have a conversion tool that we still need to drop into the Isda Create infrastructure, but we have tested that with some market participants as well,” Sloyan says.

The CDM has a model for legal documents, with a library of the components needed to represent the documents available on Isda Create. It can construct documents executed on Isda Create in digital format that conform to the CDM model.

The definition of what actually constitutes a smart contract varies, but in addition to being a self-executing contract, with its terms written in code, such contracts are often backed by distributed-ledger technology (DLT).  

Sloyan says Isda member companies are building solutions on DLT for derivatives markets that use the CDM components and library, but adds that some of these have fairly limited use cases, such as calculating the performance of an interest-rate swap and making payments, executing a securities trade, and moving securities between different custodian accounts.

Some existing services already resemble aspects of a smart contract. Richard Barton, head of product management at AcadiaSoft, says Isda Create allows the specification of different parameters and elections that can then be extracted as data.

AcadiaSoft, which is a partner of Isda Create, stores the data within its hub service, Agreement Manager. Barton says the different services his firm has around agreement data, such as initial margin calculators, threshold monitoring, and margin calculators, use that data and automate processes defined in the agreements electronically. He says these contracts and the digitization process around them are examples of a smart contract.

In DLT, there is a concept known as oracles, which are third-party services that act as middleware between the blockchain and the real world, responding to events that happen throughout the lifecycle of trades or agreements.

“We take that data, we store it, and our services around this data are really like these oracles,” Barton says.

Oracles consume inputs about daily events, process that data, and create outputs. Equally, he says, AcadiaSoft values any executed trade and aggregates it with all other trades related to that agreement. The aggregated trade triggers a margin calculation, using the captured agreement data.

“The definition of smart contracts as being self-executing and capturing all that information, that is what our services are doing,” Barton says. “A legal agreement comes out of Isda Create or offline negotiation as a PDF document, but what’s underneath it is this data.

“The bit that is missing and needs to evolve, [is that now] we have the data, but the data we are capturing are often labels for certain things, which our services have to understand,” he continues. “A true smart contract is more [about] defining the underlying logic than capturing the common name for that logic definition.”

A Smarter Future

However, a senior executive at one technology vendor says smart contracts have a long way to go before they are fully mature. Banks can more rapidly adopt new standards between themselves for processing models, but it is much harder to translate that to the entire industry. The bulk of contracts involve buy-side firms, many of which can be slow to adapt, while legacy systems and workflows also hinder adoption.

Though the CDM could play a significant role in the development of derivatives smart contracts, the project currently does not enjoy widespread adoption. The vendor executive says it can be costly to switch to a new data model.

“It’s not like you have an underlying software platform that is very flexible and you just switch it to the new data model. Data models are intrinsic to each software platform. So in order to introduce a new one, you have to put a layer on top of that. When you’ve got hundreds of systems, it is very expensive to do,” the executive says.

Unless there is a trigger—such as a regulation, or some kind of central utility emerges to unite market participants around a common standard—take-up of the CDM is likely to remain slow.

However, Isda’s Sloyan says widespread adoption of the CDM may not be necessary as far as smart contracts are concerned.

“If you take the components that are needed to build an automated solution for some part of your business, and those components are robust and referencing the right legal standards provided by Isda, you have saved yourself the effort needed to design and build a data model for that process and then confirm that it works with the legal requirements, because you have used CDM components off the shelf instead,” he says.

As far as exploiting the machine-executable components to build smart contracts, he says the core idea is that a smart contract is a bilateral agreement between two parties, or between one party and its many customers.

“So market participants don’t need to adopt the CDM, per se. They just need to say, ‘I’m happy to work with you or do business with you,’ and the smart contract is easier to build and align with Isda legal documentation,” Sloyan says.  

He admits that to pull out an internal data model and infrastructure and change it to another one can be expensive. But he says connecting to the outside world using the CDM and building mapping from internal systems to the CDM is totally feasible. 

Similar to exercises involved in mapping from Isda Create to the CDM format, a firm can do the same for its internal format to CDM, he says. The translation codes can sit in an API, which can all be in the cloud. The data can be converted or integrated internally as required.

“So, while pulling out a current data model and replacing it can be costly, this is required over time when new platforms are built for different trading functions or post-trade areas. But just comparing your data model to the CDM is very feasible in a short period of time, and not particularly costly in the grand scheme of things, considering the costs of regulatory reporting and other functionality that may be in the Isda CDM and can be utilized when comfortable with the standard,” Sloyan says.

Ciaran McGonagle, assistant general counsel at Isda, says any movement in the markets Isda covers requires some kind of a driver to lead adoption and, in recent times, these drivers tend to be regulatory in nature. When it comes to smart contracts, he says, a regulatory reporting use case might be something that is more viable in the short term, versus putting a whole trading portfolio on a blockchain and automating everything.

McDonagle says different applications will be developed over time to target specific use cases. “I think that if firms are to realize the longer-term efficiency and cost-saving benefits of new technology, a more strategic mindset is required, rather than simply implementing tactical improvements to existing processes and infrastructure,” he says.