Use Case | FRTB
TickSmith’s Data Pooling Platform empowers Canadian banks to anonymously share data for FRTB compliance
CanDeal, a leading provider of electronic marketplaces and data services for Canadian dollar debt securities and derivatives, commissioned TickSmith to provide a data pooling platform that would ensure FRTB compliance for its stakeholders*. CanDeal is the neutral and joint representative for its stakeholders, which include BMO Nesbitt Burns Inc., CIBC World Markets, National Bank Financial Inc., RBC Capital Markets, Scotia Capital, TD Securities and TMX Group. With the FRTB compliance deadline looming on the horizon in January 2022, one of CanDeal’s objectives is to reduce the cost of capital set aside for at-risk assets for the six major Canadian banks.
Regulated by the Basel Committee on Banking Supervision, the Basel III FRTB (Fundamental Review of the Trading Book) framework requires that all financial institutions globally have sufficient capital to cover potential losses on their trading book. In addition, each firm must accurately model the risk assessment using quality data in their risk models. By pooling their data into a broader and more comprehensive dataset, each firm’s risk models are likely to be more exact, which can greatly reduce capital costs.
Another objective, aside from privately pooling data, was to allow these banks to better control and monetize the combined data. The resulting derived content would not only help the banks comply with FRTB regulations, but would also allow them to generate revenue by monetizing the pooled data that they previously gave away to the markets.
TickSmith’s Data Pooling Platform had the full capabilities CanDeal needed to establish FRTB compliance for the six major Canadian banks, plus additional dataset modules for new use cases along the way. Initially, the primary requirement CanDeal was addressing was a data pooling system for FRTB. To accomplish that, each bank would need access to a shared platform to normalize and collect their proprietary data securely at scale, and to receive derived content from the collective data pool. When the data monetization use case quickly became another requirement, TickSmith’s modular system enabled the platform to adapt.
“To get the banks to approve this, I had to demonstrate that we could take the data, pool it, apply some FRTB risk qualifications, entitlements, encryption, monitoring, and an audit trail… and TickSmith had a lot of those out of the box. So for time to market, it made a lot of sense to leverage that rather than build something new,” says Andre Craig, head of DataVault Innovations at CanDeal. “The regulatory capital savings could be in the billions of dollars, so that makes the banks more efficient, and delivers meaningful cost savings.”
“We have the experience, we have both the technological know-how, as well as the regulatory know-how,” says Jimmy Huang, Data Pooling SME at TickSmith. “Regulatory compliance meets big data technology – these are very niche intersections. In terms of that specialty, we’re the only company that is able to create this distinguished product.”
TickSmith implemented their Data Pooling Platform along with the FRTB Modelability module on AWS Cloud technology for easy and fast deployment. Building on premises would not be feasible, as cloud technology allowed each bank to sign into the encrypted data pooling web portal. The workflow technology of the Data Pooling Platform normalizes and standardizes each firm’s data, yet at the same time, maintains the individual bank’s privacy and anonymity.
The platform’s enhanced entitlement capability allows participating banks to control how much data the other firms can view and access. The process of sending large amounts of data in and out and mapping them together required security measures that must be rock solid and bullet-proof. “Each bank has their own level of security measures and InfoSec requirements that we had to meet,” adds Huang. “We had to jump through each bank’s information security hoops and AWS helped us to meet those needs.”
Another challenge is that some of the OTC asset classes are quite complex and there are fewer standards. Therefore, creating common data models needed to be more involved. “We had to adapt our Analytics Platform to cope with the rules of FRTB,” says Francis Wenzel, TickSmith’s CEO. “And we had to do enhanced ETLs [extract, transform, load processes] on AWS Cloud for those asset classes.”
The data pooling solution required TickSmith to break new ground in big data technology. First, in handling millions of OTC instruments from different sources and secondly, for producing collective, consolidated data for modelability. “In Canada, a lot of instruments are traded by local institutions- so they made the case that if you could pool that data together… you can use the modelability rules to create synthetic pricing from the orders of different participants,” says Wenzel, “for FRTB, the more data you have, the better your risk models are.”
Thanks to the pooled data, each bank is able to detect continuity of observable prices for more securities or risk factors. This makes it easier for the banks to leverage their capital. Combined with the additional data monetization use case, CanDeal was able to offer a platform that would save and make the banks money; a complete win-win situation.
“The consolidation of this OTC multi-asset class data is without peer in the domestic market, which is why we’re so proud to have been asked to build the technology to support it,” continues Wenzel. “With our experience in the capital markets, we’re confident that we can deliver an unmatched service to DataVault Innovations, that will be robust enough to support an array of data, services and clients for years to come.”
*As of publishing, there are currently six Canadian banks that are stakeholders. CanDeal is in discussion with other banks interested in becoming members.