Sahm Rule vs Initial Claims
The Sahm Rule reads 0.27 percentage points in February 2026, well below the 0.50 trigger after the July 2024 cross-over that turned out to be a false positive. Initial jobless claims (FRED ICSA) for the week of April 18, 2026 printed 214,000 with a four-week moving average of 210,750, near multi-decade lows.
Also known as: Sahm Rule Recession Indicator (Sahm rule, recession indicator, Sahm) · Initial Jobless Claims (jobless claims, initial claims, unemployment claims)
Why This Comparison Matters
The Sahm Rule reads 0.27 percentage points in February 2026, well below the 0.50 trigger after the July 2024 cross-over that turned out to be a false positive. Initial jobless claims (FRED ICSA) for the week of April 18, 2026 printed 214,000 with a four-week moving average of 210,750, near multi-decade lows. The pair links a smoothed unemployment-rate signal (which has called every US recession since 1953 except 2024) with the highest-frequency labor-market data available, weekly Department of Labor administrative filings. The 2024 false positive has changed the institutional treatment of both: the BLS Beige Book and the Cleveland Fed labor-market dashboard now flag claims-Sahm divergences explicitly, because the 2024 episode showed that an unemployment rate can rise on labor-supply expansion rather than layoffs, breaking the single-variable Sahm signal in ways the dual reading would have caught.
What each indicator measures and where the data comes from
The Sahm Rule is a recession-dating heuristic developed by Claudia Sahm in 2019 while she was a section chief at the Board of Governors. It triggers when the three-month moving average of the seasonally-adjusted U-3 unemployment rate (BLS series UNRATE) rises 0.50 percentage points or more above its trailing twelve-month minimum. The Federal Reserve Bank of St. Louis publishes a real-time series, SAHMREALTIME, on FRED that uses the contemporaneous data vintage rather than revised figures.
Initial jobless claims are administrative filings collected weekly by state labor agencies and aggregated by the Department of Labor. The Employment and Training Administration releases the seasonally-adjusted national figure every Thursday at 8:30am ET, covering the prior week (Sunday-Saturday). FRED carries the series as ICSA. The data is not survey-based, which is part of why claims are considered the cleanest weekly read on actual layoff dynamics: every filing is an administrative event, not a respondent's interpretation of their employment status.
Track records: 1953-2024 for Sahm, 1967-2026 for claims
The Sahm Rule has fired ten times since 1953. Seven coincided with the start of an NBER-dated recession within one quarter (1957, 1960, 1969, 1973, 1980, 1981, 1990). Two more (2001 and 2008) fired one to three months after
90-Day Statistics
No data available
Explore Each Metric
Related Scenarios & Forecasts
Get daily macro analysis comparing key metrics delivered to your inbox. Stay ahead of market-moving divergences.
Frequently Asked Questions
What is the latest Sahm Rule reading?+
The Sahm Rule reads 0.27 percentage points as of February 2026 from the FRED real-time series SAHMREALTIME. The reading is well below the 0.50 trigger threshold and reflects the normalization of the unemployment rate after the July 2024 cross at 0.53 that turned out to be a false positive. The current 0.27 sits in the bottom decile of the 1953-2026 distribution.
What is the latest initial claims level?+
Initial jobless claims for the week ending April 18, 2026 printed 214,000 (up 6,000 from the prior week, slightly above the 211,000 consensus). The four-week moving average was 210,750. The level is among the lowest absolute readings since the 1960s and is well below the 350,000-450,000 pre-recession warning band. Continuing claims at 1.78 million are also low.
Why was the 2024 Sahm signal a false positive?+
The trigger was driven by labor-force expansion rather than labor-demand contraction. The Congressional Budget Office estimated that net immigration rose from 110,000 per year on average through 2020 to 2.4 million in 2023 and 2024, which mechanically raised the unemployment rate by an estimated 10-15 basis points as new entrants searched before finding work. Initial claims never confirmed: in every prior cycle where Sahm fired and a recession followed, the insured unemployment rate rose 47 basis points year-over-year by the cross. In July 2024 the IUR was flat and at 1.2%.
Did claims confirm or reject the 2024 Sahm trigger?+
Related Comparisons
Explore Across Convex
Data sourced from FRED, CoinGecko, CBOE, and other providers. This page is for informational purposes only and does not constitute financial advice. Past performance does not guarantee future results.