An additional explanatory variable in yield-curve recession models capturing how far the short-term policy rate sits from its recent path or estimated neutral level. Added to the term spread, it sharpens the model's signal by distinguishing genuine inversions from those driven by an unusually high front end.
How it works
Classic recession-probability models regress future downturns on the term spread (e.g. 10Y minus 3M). A short-rate deviation predictor adds the gap between the current short rate and a benchmark — its trailing average or a neutral-rate estimate — as a second regressor. The intuition: an inversion driven by a restrictive front end carries different information than one driven by collapsing long yields, so conditioning on the short-rate level improves in-sample fit and out-of-sample discrimination.
Why it matters now
After the deepest US curve inversion in four decades failed to deliver a 2023–24 recession on schedule, forecasters have leaned on augmented specifications to reconcile the signal — making short-rate-conditioning variables central to the 2025–26 debate over whether the term-spread model is broken or merely needs the policy-rate context added back in.
Example
In a modified recession specification, a model relying on the term spread alone might assign a 30–40% twelve-month recession probability; the briefing cited a version where the implied probability rose to roughly 75% once a short-rate deviation predictor was added — illustrating how conditioning on the level and trajectory of the front end can materially shift the model's output from the same yield curve.
Frequently asked
- What is a short-rate deviation predictor?
- A short-rate deviation predictor is an explanatory variable in yield-curve recession models that measures how far the short-term policy rate sits from its trailing path or estimated neutral level. Added alongside the term spread (e.g. 10Y minus 3M), it conditions the recession signal on front-end context, distinguishing inversions driven by a restrictive policy rate from those driven by collapsing long yields.
- How does a short-rate deviation predictor improve recession models?
- It improves recession models by adding a second regressor that captures the level and trajectory of the policy rate, not just the slope of the curve. An inversion caused by a high front end carries different information than one caused by falling long yields, so conditioning on the short-rate gap sharpens both in-sample fit and out-of-sample discrimination between genuine and benign inversions.
- Why does the short-rate deviation predictor matter in 2025-26?
- It matters because the deepest US curve inversion in four decades failed to produce a 2023-24 recession on schedule, forcing forecasters to ask whether the term-spread model
Glossary · term-spread model
A recession-probability model that maps the slope of the Treasury yield curve — typically the 10-year minus 3-month or 10-year minus 2-year spread — onto the likelihood of a downturn over the next twelve months, usually via a probit regression. An inverted spread signals elevated recession risk.
Read more →
- How does a short-rate deviation predictor differ from the term spread?
- The term spread measures curve slope — long yield minus short yield — while the short-rate deviation predictor measures the front end's distance from a benchmark like its trailing average or neutral rate. The spread captures shape; the deviation captures level. Used together, they let a model separate an inversion's cause, which a slope-only specification cannot do.
- Can a short-rate deviation predictor change a recession probability?
- Yes — adding the variable can materially shift a model's output from the same yield curve. In one cited modified specification, a term-spread-only model assigned a 30-40% twelve-month recession probability, while the augmented version implied roughly 75% once the short-rate deviation predictor was included, showing how front-end conditioning reshapes the forecast.