Jacob Rank-Broadley explores the current state of play in the Fundamental Review of the Trading Book (FRTB) regulation and how sourcing the right data from the right vendor will be key to regulatory compliance.
- The Basel Committee on Banking Supervision (BCBS) is using the FRTB regulation to tighten the rules on how banks measure their market risk exposure.
- The LIBOR transition and COVID-19 risk further complicating the introduction of FRTB.
- FRTB rules were finalized in January 2019. However, some jurisdictions still have much work to do in order to finalize the technical details.
For more data-driven insights in your Inbox, subscribe to the Refinitiv Perspectives weekly newsletter.
Among the myriad incoming regulations facing financial institutions, few will have a greater impact than the Fundamental Review of the Trading Book (FRTB). The Basel Committee on Banking Supervision (BCBS) is tightening the rules on how banks measure their market risk exposure, which could result in higher capital requirements for banks.
FRTB: A brief recap of the market risk rules
Banks must now not only adopt the regulator’s standardized approach, but they also have the option to design their own models for calculating their market risk exposure.
The regulation tightens up the market risk rules across the board by refining what is in scope of the trading book, putting in place stricter requirements on the quality of data and calculations for those banks seeking to use their own models, and changing the capital calculations themselves in order that they better take into account tail risk and variable liquidity horizons.
An industry problem
Banks seeking to use their own models under the internal models approach (IMA) will need to conduct a new liquidity test to demonstrate sufficient trading activity associated with their model inputs.
As banks try to pass this new liquidity test, they’ll face four key challenges:
- Many markets lack sufficient transparency from which to identify usable executed trades and committed quotes. While initiatives like MiFID II can help, there is still much work to collect the data.
- A very large number of trades and committed quotes need to be analyzed for inclusion into this test and hence sophisticated (often cloud-based) tools are required to store, process and analyze very large datasets.
- Data needs to be consolidated into a standardized data model to allow aggregation based on instrument attributes. This requires complex mapping rules to align inconsistent data models across many sources and products. For example, different taxonomies or naming conventions may be used in different asset classes or sources, and even something as basic as the date of a transaction could be problematic, since 5pm in New York is 7am the next day in Tokyo.
- Users need to be able to rapidly assess the benefit of adding different sources and refining their mapping rules.
Many banks will likely use internal data as their primary source for this liquidity test, but this won’t be enough to consistently pass the new test. Banks will need to source additional data from vendors that can offer comprehensive historical data.
Synchronizing data across internal and vendor sources can be challenging but working with a credible provider can reduce the burden.
FRTB and market uncertainty
FRTB is being implemented across a backdrop of structural change driven by the LIBOR transition and volatility relating to the COVID-19 crisis and the world’s economic response.
The LIBOR transition will be responsible for a profound shift in market liquidity as firms replace their LIBOR referencing derivatives with alternative risk free rates such as SONIA, SOFR and €STR.
As activity starts to move away from LIBOR, we’ll observe more non-modellable risk factors. While ultimately this activity should start to show up in derivatives referencing risk free rates there are two challenges.
Firstly, during the transition period liquidity is bifurcated, potentially resulting in both the LIBOR and risk free rate risk factors having insufficient trading activity to pass the test.
Secondly, since the test is conducted over a one-year period, there is a delay between liquidity picking up and a risk factor passing. If the LIBOR transition takes place swiftly and is completed by the end of the year, this shouldn’t be too much of a problem as liquidity in risk free rates will be well established before FRTB comes into effect. However, there is potentially greater concern in the U.S. if the LIBOR transition drags on into 2023.
The market correction that took place in March last year was a real test for FRTB’s new liquidity assessment. Many firms were concerned that the plunge in equity markets could be followed by paralysis of capital markets that would leave many risk factors becoming non-modellable for a full year.
Refinitiv’s analysis highlighted that for many market activity remained buoyant and so, despite the extreme conditions, the test remained valid.
Despite the global text being finalized two years ago, there is still further work required to set the technical details at a local level. Some jurisdictions such as the EU have already made significant progress by proposing draft technical standards, while other jurisdictions still have much work to do.
Defining technical standards at a local level risks divergence of the rules across jurisdictions.
One area of significant uncertainty is the definition of committed quotes. The BCBS specifies “a price from an arm’s length provider at which the provider of the quote must buy or sell the financial instrument”.
However, the EU has taken a different view and is currently proposing that quotes should have both a bid and an offer on the same day. This definition is inconsistent with market convention in some products. For example, many traditional over the counter (OTC) markets use a request-for-quote (RFQ) protocol, in which customers can specify whether they are seeking to buy or sell the instrument, and dealers respond with a single-sided quote.
Another area of uncertainty is the treatment of the capital floor, which could limit the benefits of implementing IMA. Here the EU is contemplating a potentially weaker interpretation of the rule that may benefit IMA approved banks.
Finally, there remains limited clarity on how evidence of transactions or committed quotes used in RFET should be made available to supervisors.
What are the next steps?
The EU is hoping to finalize the IMA technical standards in Q1 before entering the Official Journal by the middle of 2021.
The regulation would then kick in three years later, in mid-2024. However, before then the standardized reporting requirements will come into effect, meaning that many firms are seeking to get the basics up and running prior to the September 2021 deadline.
By contrast in the U.S., we still don’t have draft domestic rules.
Hopefully 2021 will be the year we finally see them before the Fed kicks off preparedness exams and conducts the model approval before a January 2023 go-live date. It’ll be key to see whether the U.S. follows the EU’s lead on topics such as committed quotes and implementation timelines or whether it stays true to the BCBS text.
Start preparing for now
As we work towards FRTB going live, there is much work to ensure compliance, particularly for banks seeking IMA approval. In relation to sourcing the data, it’s worth thinking about:
- In the short term: Have I got the data I need for the standardized approach such as index and fund constituents and weights data to support look through obligations?
- In the medium term: How am I going to source sufficient executed trade and committed quote data to avoid excess capital requirements under the internal models approach?
In both cases, the implementation can take longer than many firms anticipate, so starting earlier can avoid future compliance issues.