Sahm Rule: reading the recession signal

What is the Sahm Rule?

The Sahm Rule, developed by economist Claudia Sahm, is a real-time recession indicator based on the US unemployment rate. It triggers when the 3-month moving average of unemployment increases by 0.5 percentage point or more from the minimum of the last 12 months.

It confirms that an economic slowdown is becoming serious through the labor market. It does not measure equity valuation or instantaneous financial stress — it is a macro signal, not a direct equity signal.

Economic logic

Unemployment is a lagging indicator: it starts to rise after a recession begins. Once deterioration starts, it often accelerates — layoffs reduce consumption, which leads to more layoffs. The Sahm Rule captures this inflection point where labor market deterioration becomes self-sustaining.

Macro ≠ equities

Since 1970, the rule has detected all US recessions with zero false positives — but at equity market peaks, the normalized signal often remains low. A confirmed recession does not automatically imply an ongoing crash or a sell signal at the top.

How the signal is built

Step 1: data collection

ONELIX relies on the monthly unemployment rate published by the Bureau of Labor Statistics (BLS), available via FRED (UNRATE) since 1949. These raw data form the basis of the indicator.

Step 2: smoothing and Sahm Rule calculation

To reduce statistical noise, ONELIX calculates a 3-month moving average of the unemployment rate, then compares it to the lowest point of the last 12 months. The difference between these two values constitutes the raw Sahm Rule. When this difference exceeds 0.5 percentage point, the rule has historically signaled the beginning of a recession.

Step 3: ONELIX normalization (0–100%)

To facilitate interpretation and integration into the LIX, ONELIX normalizes the Sahm Rule on a 0 to 100% scale using a proprietary formula. The normalized indicator is displayed in the dashboard and used for LIX calculation.

Historical reading

The chart below presents the raw value of the Sahm Rule from 1949 to year-end 2025. Crossings of the 0.5 pp threshold generally coincide with recession phases — but equity markets may already have anticipated part of the shock.

Detected recessions (1973–2020)

Since 1970, the Sahm Rule has detected all US recessions with zero false positives. Examples of triggers between 1973 and 2020:

• 1973–75: signal at 0.53 pp (November 1973)
• 1980: signal at 0.67 pp (May 1980)
• 1981–82: signal at 0.57 pp (August 1981)
• 1990–91: signal at 0.50 pp (August 1990)
• 2001: signal at 0.50 pp (April 2001)
• 2007–09: signal at 0.57 pp (December 2007)
• 2020 (COVID): signal at 1.20 pp (March 2020) — exceptional and brief shock

Unlike the yield curve, which often anticipates recessions 12–18 months ahead, the Sahm Rule confirms them in real time, generally within 1–3 months of the official start.

The ONELIX reading of the Sahm Rule

ONELIX reads the Sahm Rule as a macroeconomic confirmation signal. It complements the yield curve, high yield spreads, valuation and volatility — adding a reading of the real economy and the labor market.

The normalized score transforms unemployment deterioration into an intuitive scale: 0% means very low recession risk, 100% means recession very likely. It is not a standalone equity timing tool.

Across 15 major corrections (≥ 19%) since 1962, the normalized signal showed clear detection (≥ 75%) in 13% of cases (1962, 1980), moderate signal (60–75%) in 7% (2025), and remained low (< 60%) in 80% of cases — notably 1966, 1969, 1973, 1987, 1990, 1998, 2000, 2008, 2011, 2018, 2020 and 2022. This low peak detection rate is expected: the Sahm Rule measures recession, not equity complacency at the top.

The signal becomes more useful when it converges with tight credit, extreme valuation or rising volatility — and when it confirms that an equity correction reflects real macro deterioration.

Product preview. Frozen historical capture of the ONELIX dashboard illustrating the Sahm Rule reading.

Limits and vigilance points

The Sahm Rule relies on economic data published with delay and sometimes revised.

• Lagging signal: unemployment often deteriorates after the first market signals — when the rule triggers, the market has often already corrected.
• Not a price signal: it does not directly measure equity valuation or credit stress.
• Macro ≠ equities: a confirmed recession and an equity crash do not systematically coincide at the peak.
• Revisions: macroeconomic data can be adjusted after publication.
• Exceptional shocks: COVID-19 produced an extreme signal (1.20 pp) but brief.
• Combined use: it must be confronted with rates, credit and volatility.