The Deflationary Cousin of Stagflation
This scenario is the mirror image of the [stagflation](/glossary/stagflation) case: instead of [inflation](/glossary/cpi) persisting as growth stalls, demand collapses and [inflation](/metrics/cpiaucsl) falls with it, producing a classic deflationary bust with widening credit spreads and a [flight to safety](/glossary/risk-on-risk-off). The distinction matters [for](/metrics/fodsp) positioning, because a hard landing is the one regime where long-[duration](/glossary/duration) Treasuries are the trade and [inflation](/glossary/cpi) hedges fail.
The Signals Are Flashing, but Quietly
The [leading indicators](/glossary/leading-economic-index) are genuinely mixed and lean cautionary. The [Sahm](/metrics/sahmrealtime) real-time indicator is near 0.28, approaching the 0.5 level that has reliably marked past recessions, [initial claims](/glossary/initial-claims) have risen [for](/metrics/fodsp) several weeks, and the [yield curve](/glossary/yield-curve) has been through the longest inversion on record. [Consumer confidence](/glossary/consumer-confidence-conference-board) is soft and the composite [recession](/glossary/recession)-probability index has been elevated. On their own, these would argue for a rising probability.
Why the Reflation Tape Wins for Now
The problem for the bear case is that markets are not confirming it. Cyclicals are outperforming defensives by around five points, [semis](/metrics/smh) and [homebuilders](/metrics/xhb) are leading, high-[yield](/glossary/dividend-yield) credit spreads are tight near 2.80%, and breadth is broadening rather than narrowing. A hard landing does not usually begin with cyclical leadership and calm credit. The credit-impulse acceleration and the reflation read in the tape are, for now, outweighing the labour-market softening. That is why we trim the probability to 16% from a 20% base rate, while keeping it on the board, because the labour signals are real and can reassert quickly.
Scenario probabilities are computed using a Bayesian log-odds model with calibrated base rates, z-score evidence weighting on first-differences, cross-metric correlation adjustment, and simultaneous coherence enforcement. Positioning reflects directional expected value under binary resolution assumptions. Full methodology and known limitations →