Deciphering FRTB: navigating regulation, data compliance and future assurance

Hot on the heels of the European Union announcing a delay to Fundamental Review of the Trading Book (FRTB) implementation, a recent Risk Live Europe panel session, sponsored by ActiveViam, explored the challenges banks are facing with these new regulatory standards
 

The panel

• Xavier Bellouard, Managing director and co-founder, ActiveViam
• Fabio Lania, Market risk manager, Intesa Sanpaolo
• Katherine Wolicki, Global head of engagement and outreach, global benchmarking initiative, Global Association of Risk Professionals (Garp)

Experts discussed whether the FRTB will achieve its regulatory aims, along with the widely anticipated decline in banks using the internal models approach (IMA). Given recent regulator announcements, the panel also considered the future course of the FRTB. This article presents the key takeaways from that discussion.

Low IMA adoption

Designed to address shortcomings that were exposed following the financial crisis that began in 2007–08, the panellists said that, in theory, the FRTB’s more granular and prescriptive market risk capital standards meet the Basel Committee on Banking Supervision’s policy objectives and mark a step forward compared with value-at-risk (VAR) calculations.

However, very few banks are applying to use the IMA; most are adopting the standardised approach (SA) for calculating their Pillar 1 capital requirements. When asked to predict the number of banks that will adopt the IMA under FRTB, panellist responses ranged from five to 12, with major US banks expected to follow this route, compared with less than a handful of European banks.

Xavier Bellouard, ActiveViam

Panellists raised practical concerns in several areas. Katherine Wolicki, global head of engagement and outreach, global benchmarking initiative, at Garp, underlined the complexity of putting the FRTB components together and running them day to day. “This complexity is reflected in the low number of firms that are choosing to go down the IMA path,” she said. “It is really expensive, very complicated and, ultimately, pretty uncertain.”

However, Xavier Bellouard, managing director and co-founder of ActiveViam, cautioned that, if banks only adopt the SA, it could prove to be a backwards step in risk management. “The SA is no more than a pretty good parametric VAR, which means we are going to fall back into the same problems we had under Basel II. Despite the benefits of the IMA, with so few banks using it, there will be no improvement.”

The risks of decoupling

The panel discussed the risks of banks adopting the SA for their Pillar 1 capital requirements and the IMA for Pillar 2. Bellouard explained: “There is a huge risk of divergence between what banks report to the regulator and what they use as trading decision metrics or day-to-day risk controls. For the past 10 years, many banks have been trying to employ one golden source of data across trading and risk management. If we depart from that, I think it would be a huge step back.”

Fabio Lania, market risk manager at Intesa Sanpaolo, one of the few banks in the EU applying to use internal models, stressed: “Decoupling the strong link between market risk management and the capital requirement calculation at the trading level carries a big risk that I don’t think regulators are considering.”

NMRFs and P&L attribution tests

Non-modellable risk factors (NMRFs) and profit-and-loss (P&L) attribution tests were highlighted as particular FRTB challenges.

Bellouard noted that NMRF constraints are a hugely difficult issue, while Lania cautioned: “The criteria for modellability are too tight, especially in Europe. In addition, the aggregation formula is based on very strong assumptions, leading to capital numbers that are disproportionate to the real risk of the instruments they are sitting behind.”

Bellouard and Wolicki also stressed the significant data challenges banks are facing: a lack of data and poor quality data. Third-party solutions would help, however, as Wolicki underlined: “It has become a chicken-and-egg situation. There have been lots of discussions around the potential for vendor solutions. But, with so few banks adopting the IMA, there is no market for it and it is not interesting for vendors.”

Panellists discussed the importance of building a common risk factor taxonomy across finance and risk that aligns with the P&L, ensuring risk models match exposure. Bellouard stressed that cleaning up data and ensuring consistency across systems are important first steps for P&L attribution. “Having the same taxonomy and maintaining a unique, golden source of data would play a big role in improving P&L attribution data, because most errors can be fixed in those cases,” he noted.

Future uncertainties

Although regulators have not expressed a strong desire to increase IMA adoption, the panel agreed that the more complicated elements of the FRTB package would need revisiting. Bellouard said: “Tweaking or loosening some of the constraints would go a long way to increasing IMA adoption.”

Lania insisted this did not mean rewriting the regulation from scratch. “There are two to three parameters, such as NMRFs and P&L attribution, that they could adjust,” he said.

There are some indications that goalposts are shifting. The EU’s decision to push back implementation of the FRTB to January 2026 was driven by the need to ensure a level global playing field. Following intense opposition from banks in the US, the Federal Reserve has indicated that significant changes to FRTB are likely and there are widespread expectations the US will be unable to meet its July 2025 start date.

Lania observed: “We are looking very carefully at the US to see whether it will nudge European institutions to relax some of the FRTB constraints. We are hoping that, whatever happens, regulators act quickly. If too much time passes, it will be very hard for some banks to change back to the IMA as they would need to rebuild teams and infrastructure.”

In summary

While FRTB has long been in the pipeline, uncertainties remain over its timing and final substance, particularly in the US. IMA take-up is predicted to be very low because of the cost and complexity of these new standards, particularly with obstacles in the form of NMRFs and P&L attribution tests. Banks are also struggling with data, and there is currently little vendor appetite to plug the gap. Without a relaxation of some of the regulatory constraints, this picture is unlikely to change. However, banks will be keeping a close eye on the next moves in the US, to gauge any likely implications for the EU and UK markets.

Sponsored content