Recession Probability
The model-implied or market-implied likelihood of a recession over a forward window — typically 12 months — derived from yield-curve, credit, and labor signals.
Definition
Recession probability models combine leading indicators (yield curve, credit spreads, ISM new orders, jobless claims) into a single time-varying probability. Different models weight signals differently; the NY Fed term-spread model is the most widely cited.
No single model is reliable in isolation, which is why institutional desks track several and weight them by current regime.
Public research record
Research framework · page evidenceCompleted application: Recession Probability
Recession probability drives the price of duration, the credit spread, and the equity risk premium. Mispricing it is the largest source of cross-asset miscalibration.
| Type | Claim | Scope |
|---|---|---|
| Release fact | The model-implied or market-implied likelihood of a recession over a forward window — typically 12 months — derived from yield-curve, credit, and labor signals. | Macro |
| Release fact | The 2s10s inverted in July 2022; the NY Fed model peaked above 70% in 2023. The recession arrived later and shallower than expected — a cautionary tale about probability ≠ certainty. | Published example |
Countercase, invalidators, and sources
The relationship is conditional: another driver can dominate the same asset over the selected horizon.
- · The underlying observation changes materially.
- · The selected asset has no measurable exposure to this mechanism.
Why it matters
Recession probability drives the price of duration, the credit spread, and the equity risk premium. Mispricing it is the largest source of cross-asset miscalibration.
Worked example
The 2s10s inverted in July 2022; the NY Fed model peaked above 70% in 2023. The recession arrived later and shallower than expected — a cautionary tale about probability ≠ certainty.