Fundamental Review of the Trading Book - FRTB

The Internal Models Approach (IMA) is one of two methods banks can use to calculate market risk capital requirements under the forthcoming Fundamental Review of the Trading Book. The other is the standardised approach.

To qualify for IMA status, banks must pass two regulator-set tests: the P&L attribution test, which gauges how accurately a bank is able to measure the profits and losses a given trading desk generates; and back-testing, which compares the P&L figures with value-at-risk metrics. If a desk fails either test, it faces being relegated to the standardised approach.

Desks that opt for the IMA may face costly capital add-ons to account for non-modellable risk factors.

If a desk fails the IMA eligibility tests, it will revert to Standardised Approach with corresponding capital treatment. Significant work across Front Office, Finance and Risk has been undertaken to try and ensure that the Backtesting and P&L Attribution eligibility tests can be achieved. The quality of P&L, risk and market data is crucial.

Stringent rules on whether risk factors are sufficiently frequent / observable; with potentially significant capital add-ons if observation criteria cannot be met. The risk factor must pass a risk factor eligibility test (RFET), in which the bank must identify for the risk factor at least 24 real price observations per year on a monthly basis or at least 100 “real” price observations over the previous 12 months.

The Credit Valuation Adjustment (CVA) is a change to the market value of derivative instruments to account for counterparty credit risk. It represents the discount to the standard derivative value that a buyer would offer after taking into account the possibility of a counterparty’s default. CVA is the most widely known of the valuation adjustments, collectively known as XVA.

The offical go-live for the FRTB regulation is January 1^st 2025. However the go-live date will be preceded by a series of regulatory milestones that UBS and other banks must meet.

In June 2024, in light of the European Commission announcement on delaying the FRTB rules implementation, the Swiss Federal Council has reconfirmed that the new standards will apply in Switzerland at start of 2025, as previously communicated.

The program's overall approach has therefore been to optimise efforts where viable to get desks on IMA and manage against increased Market Risk RWA.

The capital impact of FRTB is measured through the regulatory Quantitative Impact Study (QIS). The regulatory objectives of the QIS are to monitor the size of capital impacts and finalize the calibration of the parameters governing the revised capital charge.

A QIS style analysis was performed end 2022 to assess current IMA performance against likely capital outcomes. The analysis was in-line with the 3YSP.

The Profit and Loss Attribution Test (PLAT) is a test under FRTB for assessing the robustness of a bank’s risk management models by comparing the risk theoretical P&L (RTPL) predicted by trading desk risk management models with the hypothetical P&L (HPL). 

The PLA requirements are based on two test metrics:

1. the Spearman correlation metric to assess the correlation between RTPL and HPL; and
2. the Kolmogorov-Smirnov (KS) test metric to assess similarity of the distributions of RTPL and HPL.

Based on the outcome of the metrics, a trading desk is allocated to either a red, amber, or green zone.