Skip to main content
2diff.io

Econ Blog

rates
By Second Difference SolutionsJanuary 28, 2026Updated October 9, 2026Data as of October 8, 2026

10Y-2Y Treasury spread

The 10-year Treasury yield minus the 2-year Treasury yield, tracked since 2018. When it inverts, recession has historically followed. The chart shows the inversion window and what came after in this cycle.

The 10-year Treasury yield minus the 2-year Treasury yield ("the spread" or "the curve") is one of the most-watched macroeconomic signals. When it inverts (2-year yields more than 10-year), recession historically follows within 6-18 months. The 2022-2024 inversion broke the streak.

26 months
Inversion duration

Jul 2022 to Aug 2024

Longest
Inversion on record

FRED history from 1976; the prior record ran 1978 to 1980

0 of 1
Recession hit rate

In this cycle

10-year Treasury yield minus 2-year Treasury yield, 2018-2026
Loading chart…

The chart crosses zero into negative territory in Jul 2022, troughs at -1.08% in Jul 2023, and re-emerges above zero in Aug 2024. Whether the inversion 'worked' as a recession signal in this cycle is the most-debated question of the 2020s recovery.

The historical track record

Since 1980, every recession has been preceded by a 10Y-2Y inversion:

  • 1981-82: inversion 1980; recession started July 1981
  • 1990-91: inversion 1989; recession started July 1990
  • 2001: inversion early 2000; recession started March 2001
  • 2007-09: inversion 2006; recession started December 2007
  • 2020: brief inversion August 2019; recession started February 2020 (pandemic-induced; arguably coincidental)

Five for five if you count 2020, four for four if you don't. That's a near-perfect signal, until 2022-24.

What happened in 2022-24

The 10Y-2Y spread inverted in Jul 2022 and stayed negative through Aug 2024: 26 months, the longest run since the 1970s. By the historical 6-18 month lag, a recession should have started by January 2024. None did.

Three explanations are debated:

  • Coincidental track record. The historical correlation was always smaller-sample than it looked. With only ~5 events in 50 years, the signal could have been noise that happened to align with recessions.
  • Post-COVID structural shifts. Fed balance sheet expansion, labor market tightness, and fiscal stimulus altered the transmission mechanism between yield-curve signals and real-economy outcomes.
  • The recession is delayed, not avoided. Even in 2026, some economists argue the 2022 inversion's recession signal is still pending, just shifted from the historical 6-18 month lag to something longer.

Each has supporters. None has been proven.

Practical implications

For the 2-year-ahead investor, the question becomes: how much weight to put on yield-curve inversions in a recession-probability model?

Want this kind of read on your own business? Try the BizAnalyzer demo on sample data, no sign-in needed.

3 posts