The Yield Curve and Recession Signals

What the curve tells you

Plot government bond yields by maturity and you get the yield curve. Normally it slopes up — longer lending commands more compensation. When short yields exceed long yields, the curve is inverted, and that has historically preceded recessions.

The two spreads that matter

• 2s10s — the most quoted. Two-year versus ten-year.
• 3m10y — the measure with the strongest historical record, and the one some central banks reference explicitly.

Why inversion predicts

Inversion usually means the market expects the central bank to cut in the future — because policy is restrictive enough to slow the economy. It is a statement about the expected path, which is why it connects so directly to everything in Part 2.

The brutal caveat: timing

Inversion has preceded every recent recession, but the lead time has ranged from several months to more than two years. During that gap, risk assets have often rallied substantially. If you short equities the day the curve inverts, you can be right about the recession and still be stopped out twice before it arrives.

Un-inversion is the real warning

Historically, recessions have tended to begin after the curve steepens back out of a deep inversion — when the front end falls as the central bank starts cutting in response to visible weakness. The steepening is closer to the event than the initial inversion.

Use it as a risk dial, not a trigger

The practical use is not "inversion means short everything now". It is: when the curve is deeply inverted, reduce risk, avoid marginal longs, and be quicker to take profits. It is a background condition, not an entry signal.

Other signals worth watching alongside

• Credit spreads — corporate bonds widening versus government bonds shows stress in funding markets.
• Leading indicators — composite leading indicators bundle several series; they have a mixed record on timing.
• The unemployment rate's slope — the Sahm rule uses a small rise in unemployment to flag early recession.

How to check this yourself

Look up the two-year and ten-year yields and subtract. Note the sign and roughly when it flipped. Then find when the last three inversions began and how long it took before anything actually broke. The signal is real, and it is also very slow — which is the part every headline leaves out.