Severe Middle East scenarios have weaker historical precedent
The European Central Bank (ECB) finds that severe Middle East conflict outcomes have less historical precedent than milder alternatives in its September 2026 projections. Its toolkit scores growth and inflation paths, not the probability of conflict events.
Milder paths score higher
Applied to the September 2026 ECB staff projections, the framework assigns the milder Middle East scenario empirical plausibility scores of 82-96 and the adverse scenario 56-85. The severe scenario scores lower across methods but remains relevant for policy risk assessment.
Cumulative absolute deviations from baseline range from 0.2 to 1.2 percentage points for real GDP growth and 1.0 to 4.3 percentage points for HICP inflation.
Scenarios differ in energy prices, financial uncertainty and indirect and second-round inflation effects.
On a scale from 0 to 100, zero indicates virtually no historical precedent; 100 means paths are as plausible as the baseline.
The scores assess joint economic outcomes, not the likelihood of the events behind them.
Dependencies change the verdict
Bayesian quantile regressions and copulas construct joint distributions using historical data and the ECB Survey of Professional Forecasters.
Baseline-centred densities use 20-year rolling windows starting in the early 1980s; each scenario receives its highest relative likelihood across windows.
For 2026-2028, the adverse scenario scores 55.7 against 0.1 for a similarly severe flat-path scenario when both temporal and cross-variable dependencies count.
Ignoring either dimension can produce misleadingly high scores.
Precedent is not probability
The ECB cautions that low scores need not invalidate well-founded scenarios whose outcomes lack historical precedent.
The metric cannot separate shocks, transmission channels and model misspecification.
Expert and policy judgement therefore remain necessary alongside the scenario narrative.
IN: