The US recession prediction market has been one of the most interesting trades of the 2020s. Economists called for recession after the Fed's aggressive rate hike cycle. The recession didn't come. Then they called for it again. It still didn't come. Now the 'soft landing' has become the consensus narrative, which — if you know anything about markets — is usually the moment to start pricing in the alternative.
What Defines a Recession Market
The standard recession definition is two consecutive quarters of negative GDP growth, though the NBER uses a broader set of indicators. Boromarket's recession prediction markets track both: a simple '2 consecutive negative GDP quarters by X date' and a more nuanced 'NBER officially declares recession by X date' — the latter often trails the former by 6-18 months.
- →Leading indicators: yield curve inversion, PMI data, consumer confidence
- →Coincident indicators: GDP, industrial production, employment
- →Lagging indicators: unemployment rate, credit delinquencies
- →Sahm Rule: when unemployment rises 0.5% from its low, recession is typically underway
- →Soft landing definition: inflation falls to target without significant job losses
The Perpetual Bear Problem
There's a well-documented phenomenon in recession prediction markets: perma-bears get a lot of attention and very little accuracy. Being wrong about recession for three years and then right in year four doesn't mean your model works — it means you were wrong repeatedly and eventually got lucky. Boromarket's calibration scores ruthlessly expose this. Past accuracy, tracked over time, is the only thing that matters.
US recession markets are a long-cycle play. The question isn't whether recession will ever happen — it will — it's whether your timing model is actually calibrated or just persistent.