Market risk limits are one of the most important control mechanisms in any bank's markets business. They translate the institution's overall risk appetite — as approved by the board — into specific, measurable constraints on how much market risk each trading desk, each trader, and each product area is permitted to carry. Understanding how limits are defined, how they cascade through the organisation, and how breaches are handled reveals a great deal about the risk culture and governance of a markets business.
Types of Market Risk Limit
No single risk metric captures all dimensions of market risk. A well-designed limit framework uses multiple complementary measures, each capturing a different aspect of the risk profile:
Value at Risk (VaR)
VaR is the most widely used portfolio-level risk metric. It estimates the maximum loss that a portfolio would not be expected to exceed over a given time horizon (typically one trading day) at a given confidence level (typically 99%). A desk with a daily 99% VaR of £5 million would not expect to lose more than £5 million on more than 1% of trading days — in other words, roughly two to three times per year. VaR is useful as a summary measure of overall risk, but it has well-known limitations: it says nothing about losses beyond the confidence threshold (tail risk), it assumes normal market conditions, and it can understate risk when correlations break down in a crisis. Banks supplement VaR with Stressed VaR (calculated using a stressed historical period) and Expected Shortfall (which averages losses beyond the VaR threshold).
DV01 (Dollar Value of 01)
DV01 is the change in value of a rates position for a one basis point (0.01%) parallel shift in interest rates. A desk with a DV01 of +£100,000 per basis point would gain £100,000 if rates fell by one basis point and lose £100,000 if rates rose by one basis point. DV01 limits are typically set by tenor bucket — a separate limit for the 2-year bucket, the 5-year bucket, the 10-year bucket, and the 30-year bucket — to prevent risk being concentrated in a single part of the curve even if the total DV01 is within limits.
CS01 (Credit Spread 01)
CS01 is the credit equivalent of DV01 — the change in value of a credit position (credit default swap, corporate bond, or credit-linked note) for a one basis point move in credit spreads. A credit desk's CS01 limit controls how much directional credit risk the desk can carry, and may be further subdivided by rating bucket (investment grade vs high yield), sector, or geography.
Vega
Vega measures the sensitivity of an options book to changes in implied volatility. A desk with a large long vega position benefits from increases in implied volatility; a short vega position benefits from volatility falling. Vega limits are important for options desks in equity, rates, FX, and commodities, and are often structured by maturity bucket and strike level to capture the shape of the volatility surface exposure.
Notional Limits
Despite the sophistication of risk-sensitive measures, notional limits — simple caps on the total face value of positions in a given instrument or sector — remain a useful blunt instrument. Notional limits are easy to monitor and hard to game, and they provide a check on very large positions that might not be fully captured by model-based risk measures (for example, because the underlying liquidity assumptions in the VaR model are overly optimistic).
How Limits Are Set: From Risk Appetite to Trader Limits
The limit-setting process cascades from the top of the organisation downward:
- Board risk appetite: The board approves an overall risk appetite statement, expressed in terms of maximum VaR, maximum stressed loss, and qualitative risk tolerance statements. This is the ultimate constraint within which all market risk must be managed.
- CRO allocation to business lines: The Chief Risk Officer allocates the overall VaR and sensitivity budget across the bank's major business lines — rates, credit, FX, equity, commodities — based on the strategic plan, the historical contribution of each business to revenue, and the capital available to support each activity.
- Desk limits: Within each business line, the Head of Risk for that area allocates limits to individual desks — for example, within rates, separate limits for the government bond trading desk, the swaps desk, the inflation desk, and the repo desk.
- Trader limits: Within each desk, the Head of Trading allocates limits to individual traders or books. A junior trader will have significantly smaller limits than a senior trader, and limits are typically reviewed and adjusted annually as part of the performance and risk review process.
Limit Monitoring: Real-Time vs End-of-Day
Risk limits must be monitored against live positions throughout the trading day, not just at end-of-day. Real-time limit monitoring systems consume live trade feeds from the front-office trading systems and update risk metrics continuously as new trades are booked or market prices move. This allows the market risk team to identify potential limit breaches intraday and intervene before the close.
End-of-day monitoring produces the official limit utilisation reports used for management reporting and regulatory purposes. These are based on the fully reconciled P&L and risk positions after Trade Support has resolved any booking errors or discrepancies identified during the day.
Breach Escalation Process
When a limit is breached, a defined escalation process is triggered. Minor breaches — where utilisation is marginally above the limit — may be resolved quickly by the trader reducing risk, with the market risk desk documenting the breach and the action taken. More significant breaches require escalation to the desk head and the market risk manager, with a formal explanation from the trader of why the breach occurred and how they intend to bring the position back within limits. Severe or persistent breaches must be escalated to the CRO and, depending on materiality, to the Risk Committee. All breaches are documented, and the aggregate breach history is reviewed by senior management and the board's risk committee as an indicator of the effectiveness of the limit framework.
Temporary Limit Increases
There are legitimate business situations where a temporary limit increase — above the standing approved limit — is appropriate. A trader may need to accommodate a large client transaction that would take the desk briefly above its DV01 limit, with the intention of hedging the incremental risk promptly. Temporary increases must be approved by the appropriate authority (typically the CRO or designated delegate, depending on the size and duration of the increase), documented, and subject to time limits. The frequency with which temporary increases are requested and granted is an indicator of whether the standing limits are appropriately calibrated to the business's actual activity level.
Limits as Risk Culture Signal
The way a markets business treats its risk limits reveals a great deal about its risk culture. A business where limits are consistently adhered to, where breaches are promptly reported and resolved, and where limit increases go through a rigorous approval process has a healthy risk culture. One where limits are routinely breached with minimal consequence, where the risk function is treated as an obstacle rather than a partner, or where limit frameworks lag the actual risk profile of the business is far more vulnerable to large unexpected losses.