Charta is on Google Play: practise on the go. Get the app
Charta

Explain how the FRTB amends Basel rules for deciding between trading and banking book, for backtesting, for P&L attribution, for credit risk, and for securitizations.

FRM® Part II · Market Risk Measurement and Management · Fundamental Review of the Trading Book · FRM2.MR.18.c

The idea

FRTB tightens several Basel rules. The trading book boundary moves from stated intent to a presumptive list of instruments that belong in the trading book. Switching books is allowed only in extraordinary circumstances with supervisory approval and no capital relief. Backtesting and P&L attribution are run desk by desk. A desk fails backtesting if over the latest 250 days it has more than 12 exceptions at the 99 percent VaR level or more than 30 at 97.5 percent. The P&L attribution test compares hypothetical P&L with risk-theoretical P&L from the risk model. A failing desk must use the standardized approach. For credit risk, the incremental risk charge becomes a default risk charge, a 99.9 percent one-year VaR, while spread risk sits in ES. Securitizations cannot use internal models and go to the standardized approach. Exam point: approval, backtesting and attribution are tested at desk level, not bank level.

A practice question

From the question bank, with the reason each wrong option is wrong.

How does the FRTB define which instruments belong in the trading book, compared with the earlier Basel approach?

  1. It places in the trading book every instrument carried at fair value in the accountsWhy not: Accounting measurement is not the test; the boundary rests on instrument types and trading purpose.
  2. It lets each bank choose freely, as before, but requires the choice to be disclosedWhy not: The earlier approach relied on intent and was open to arbitrage; FRTB adds presumptions and supervisory oversight.
  3. It places all instruments in the banking book unless they are traded on an exchangeWhy not: No such exchange-based rule exists; positions held for short-term resale or hedging trading positions are presumed trading book.
  4. It adds a presumptive list of trading book instruments, rather than relying only on stated trading intent - correct

The earlier boundary rested on intent, which allowed capital arbitrage. FRTB adds a presumptive list and supervisory powers so that short-term trading-type positions land in the trading book.

A harder one

What does the FRTB P&L attribution test compare, and what is its purpose?

  1. Hypothetical P&L against risk-theoretical P&L, to check the risk model captures P&L drivers - correct
  2. Actual P&L against the 99 percent VaR, to count how many days losses exceed VaRWhy not: That describes backtesting, which counts exceptions; P&L attribution compares two P&L measures.
  3. The 99 percent VaR against the 97.5 percent ES, to confirm ES exceeds VaRWhy not: The test compares P&L series, not risk measures; ES and VaR are not compared.
  4. Front-office P&L against accounting P&L, to detect misstatement of earningsWhy not: Attribution is about whether the risk model explains P&L, not an audit of reported earnings.

Hypothetical P&L revalues the previous day's portfolio with front-office models; risk-theoretical P&L is computed from the risk model's factors. If they diverge, the risk model misses real drivers, and the desk may lose internal-model eligibility.

The hardest tier

A desk on the internal models approach holds a long position in a single-name corporate bond. Under FRTB, which pair of charges captures its spread risk and its jump-to-default risk respectively?

  1. The default risk charge for spread changes, and expected shortfall for jump-to-defaultWhy not: Reverses the roles: spread moves feed ES, and jump-to-default is the job of the DRC.
  2. Expected shortfall for both, since a 97.5 percent ES over a long liquidity horizon already includes defaultsWhy not: Defaults are rare and sit far beyond the 97.5 percent tail, which is why a separate 99.9 percent DRC exists.
  3. Expected shortfall with liquidity horizons, and the default risk charge measured as a 99.9 percent one-year VaR - correct
  4. The stress scenario charge for spread and the residual risk add-on for defaultWhy not: The stress scenario charge covers non-modellable factors and the residual risk add-on covers exotic payoffs, neither of which is the default risk charge.

Daily spread movement is a market risk factor in ES. Default is an extreme event captured by the default risk charge, a 99.9 percent one-year VaR that replaces the Basel 2.5 incremental risk charge and also covers equities.

In the app: 6 practice questions on this outcome, spaced review that brings it back before you forget it.

Start free →

More in Fundamental Review of the Trading Book