1. What scenario analysis is
Scenario analysis answers:
“If this specific set of market moves happens tomorrow, what is the P&L on today’s book?”
Core ingredients:
- A base snapshot of risk factors (curves, spreads, FX, vol, etc.).
- One or more scenarios: deterministic changes to those factors.
- A revaluation engine: reprice portfolio under shocked snapshot.
- An aggregation layer: sum and slice results by book, desk, factor.
Scenarios are not probabilities. They are if-then experiments.
2. Types of scenarios
Typical types:
- Simple factor shocks:
- “Rates +50 bps parallel”.
- “Winter power +20 €/MWh, summer −5 €/MWh”.
- “All FX −10% vs CHF”.
- Shape / spread scenarios:
- “Steepen curve: long end +20, short end −10”.
- “Spark spreads widen by 5 €/MWh”.
- “Basis between hub A and B doubles”.
- Composite ‘macro’ scenarios:
- “Risk-off: equities −10%, credit spreads +100bp, gas +30%, power curves +20€ on winter strips”.
Each scenario is a function:
Then you re-run valuation.
3. Aggregation and use on a desk
Outputs per scenario:
- Total P&L for the book.
- Breakdown:
- By desk / trader,
- By product type (vanilla vs structured),
- By factor set (level, shape, spreads, basis).
Use-cases:
- Risk limits:
- “In scenario S, loss must not exceed X.”
- Risk appetite:
- Board / senior management want a view under key narratives.
- Model sanity:
- Compare scenario P&L to intuition and to historical episodes.
Scenario analysis is where quant pricing meets risk governance.
4. Interview soundbite
“Scenario analysis is deterministic: we take a base snapshot, apply coherent shocks to risk factors, and revalue the entire portfolio. The result is scenario P&L, which we then aggregate by desk, product, and factor. For power/commodities, scenarios must include not just parallel moves, but curve shape shocks, spark/basis spreads, and stress on key seasons like winter strips.”