Recession Risk 34/100 — September 27, 2026
Near-term (next 90 days) recession risk is MODERATE, not elevated, because the highest-weight real-time labor trigger (Sahm Rule) is not close to firing and layoffs remain low. The Fed tightened on September 16, 2026 by 25 bps to a 3.75%–4.00% target range, which raises policy-restriction risk, but financial conditions are still not signaling imminent stress (HY OAS ~2.80% as of Sep 24). Growth nowcasts are not recessionary: Atlanta Fed GDPNow for 2026:Q3 is roughly mid-4% (most recently ~4.4% on Sep 10 / updated Sep 25). The main recession-adjacent warning is a deterioration in forward-looking labor and goods-cycle indicators (temporary help contraction, freight weakness) alongside extremely depressed consumer sentiment (UMich August 51.7; September final reported ~48.1), which increases downside tail risk if it spills into spending.
Recession Risk Score: 34/100 — MODERATE (-4 vs 30 days ago)
Today’s Recession Risk Score is 34/100 (MODERATE), and it has fallen by 4 points over the past 30 days (from 38 on Aug 28, 2026 to 34 on Sep 27, 2026). The overall macro picture still reads as late-cycle-but-expanding, with labor-market break signals not close to firing. The “stress” side of the dashboard (credit, broad financial conditions) remains calm, even as forward-looking goods/labor scouts (temporary help, freight) and sentiment continue to flash caution.
Score Trend — Last 30 Days
The last 30 days show a gentle de-risking: Start 38 → End 34 (Δ -4), with a range-bound profile (Min 34 / Max 39 / Avg 35, 31 samples). In other words, the model has been mean-reverting around the mid-30s, not trending toward a regime shift (either recessionary acceleration or “all-clear” expansion).
The shape over the last ~10 days is the key tell. The score whipsawed between 34 and the upper-30s—38 on Sep 18, then down to 34 on Sep 19–21, back up (38 on Sep 22, 39 on Sep 23, 38 on Sep 24) and then returned to 34 on Sep 25–27. That pattern is consistent with headline-driven volatility (Fed communications, oil/long yields, sentiment headlines) layered over stable underlying internals: the high-weight real-time labor triggers are not deteriorating fast enough to “stick” the score in the high-risk zone.
Bottom line: the score is not collapsing lower (because there are genuine leading warnings), but it is also not stair-stepping higher (because labor stress and credit stress remain contained).
Key Drivers
-
The labor “tripwires” are still not tripping (big anchor for MODERATE, not HIGH).
- Sahm Rule: -0.07 (SAFE) — materially below any trigger threshold, implying no broad-based unemployment acceleration signal yet.
- Initial Jobless Claims: 197K (SAFE) — the latest weekly report (week ending Sep 19, 2026) showed claims at 197,000, down 1,000 from the prior week’s revised level. (content.govdelivery.com)
-
Fed policy tightened—but markets are not yet reflecting imminent stress.
- The FOMC raised the target range by 25 bps to 3.75%–4.00% on Sep 16, 2026. (federalreserve.gov)
- That adds policy-restriction risk (especially with depressed sentiment), but it has not (yet) transmitted into broad spreads/conditions the way pre-recession phases typically do.
-
Financial conditions remain loose by the data.
- Chicago Fed NFCI: -0.56 (SAFE) — still indicating easier-than-average conditions. (fred.stlouisfed.org)
- Credit is not pricing recession: HY OAS ~2.80% (Sep 24, 2026) sits in “complacent/tight” territory versus classic pre-recession widening cycles. (ycharts.com)
-
Growth nowcasts are expansionary (removes “imminent recession” framing).
- Atlanta Fed GDPNow (2026:Q3) was 5.0% on Sep 25, 2026, down slightly from 5.1% on Sep 17. (atlantafed.org)
- That’s not a guarantee of durability, but it is inconsistent with a near-term contraction narrative.
-
Leading indicators: softening, but not collapsing.
- Conference Board LEI fell -0.1% in Aug 2026 to 99.5, after rising +0.2% in July—i.e., a first decline since spring-like momentum, but not the kind of persistent “3Ds” slide that typically accompanies recession approach. (conference-board.org)
-
The recession-adjacent warnings are real: “scouts” are deteriorating.
- Temporary Help Services: 2520K (DANGER) — staffing is often an early labor-market canary.
- Freight Transportation Index: -0.3 (DANGER) — goods-side weakness is consistent with late-cycle deceleration risk.
- Consumer Sentiment (UMich): 51.7 (DANGER), with September final at 48.1, signaling crisis-level pessimism that can turn into spending restraint if it becomes behavior (not just attitude). (sca.isr.umich.edu)
Category Breakdown
-
Primary Indicators: 3 safe / 4 watch / 2 danger
Mixed but stable: labor stress triggers are mostly not firing, while forward-looking labor texture remains soft. -
Secondary Indicators: 2 safe / 0 watch / 1 danger
A small set, but it leans “okay now, watch later”—consistent with a slowdown risk rather than contraction. -
Housing & Construction: 0 safe / 1 watch / 1 danger
Housing is the most cyclically sensitive pocket on this dashboard; it remains a drag even if it’s not collapsing. -
Business Activity: 2 safe / 1 watch / 0 danger
Business activity is not recessionary in aggregate—more “below-trend expansion” than “contraction.” -
Consumer Credit Stress: 1 safe / 2 watch / 1 danger
Stress is building at the margin (delinquencies / low savings), but not yet at “systemic tightening” levels. -
Market Signals: 6 safe / 3 watch / 5 danger
A classic late-cycle mix: index levels and volatility read “risk-on,” while valuation/proxy ratios and some cross-asset fear gauges look stretched. -
Liquidity: 0 safe / 1 watch / 2 danger
Liquidity looks less supportive; depletion in some facilities can matter if paired with a shock. -
Real-Time / High-Frequency: 0 safe / 1 watch / 1 danger
The high-frequency layer is not screaming recession, but it is not clean either—this is where turning points show up first.
Biggest Movers
-
NY Fed Recession Probability (0.9%): -88.1% (7D) — contradictory (improving)
A sharp drop is consistent with the broader message from markets/conditions: recession odds are not accelerating in real time. -
ON RRP Facility ($5B): -81.7% (7D) — confirmatory (worsening liquidity risk)
Rapid depletion can be benign (cash moving elsewhere) but reduces a “buffer” and can amplify funding sensitivity if volatility spikes. -
Housing Starts (1275K): +21.2% (7D) — contradictory (improving)
A rebound helps cushion cyclical risk, though housing remains below prior-cycle norms in level terms. -
Yield Curve (2s30s) (0.60): +9.4% (7D) — mostly contradictory (improving)
Steepening reduces “immediate” recession signal intensity compared with deep inversion regimes. -
Chicago Fed NFCI (-0.56): -4.5% (7D) — contradictory (improving)
More negative NFCI = easier conditions; this is not what you see when recession is imminent. (fred.stlouisfed.org)
90-Day Indicator Trends
Important constraint: the provided “90-day history” series snapshots shown here contain limited dates (many series stop around July 2026), so trend math is best read as directional rather than fully time-complete.
Labor & income (the recession core)
-
Sahm Rule: 0.10 (Jun 29) → 0.07 (early Jul) → -0.07 today (SAFE).
Direction of travel is clearly away from trigger, consistent with “no broad labor break.” (30/60/90-day comparisons imply improvement vs late June/early July levels.) -
Initial claims: ~215K (Jun 29) → ~208K (mid-Jul) → 197K (Sep 19 week).
Claims are trending lower, not higher. (content.govdelivery.com) -
Unemployment rate: the history block shows 4.3% (late Jun) drifting toward ~4.2% (mid-Jul); today’s dashboard lists 4.1% (WATCH).
That implies unemployment is not accelerating, though you still watch for inflection if hiring slows. -
Real personal income (ex transfers): essentially flat at $16.6T across the history window and today.
This reads like a stable income base, supporting continued consumption unless sentiment/credit clamp down.
Business cycle “scouts”
-
Temporary help services: ~2490K (Jun 29) → ~2499K (early/mid Jul) → 2520K today (DANGER).
The label implies structural weakness even if the short window shows little change; the signal matters because temp help often turns before broader payrolls. -
Freight index: 0.5 (Jun 29) → 0.3 (early Jul) → -0.3 today (DANGER).
That’s a material deterioration across the quarter: goods movement is losing momentum, consistent with a manufacturing/retail inventory caution phase.
Financial conditions & credit
-
Credit spreads (HY OAS): hovered around ~270–280 bps in the snapshot window; today reads 270 bps (SAFE) and market data shows ~2.80% on Sep 24. (ycharts.com)
Translation: credit is not confirming recession risk. -
NFCI: ~ -0.52 (late Jun) to ~ -0.54 (mid/late Jul); today -0.56 (SAFE). (fred.stlouisfed.org)
Financial conditions have eased modestly, which is anti-recession in the near term.
Growth nowcast / leading indicators
-
Conference Board LEI: level shown stable in the snapshot, while the latest release notes -0.1% m/m in Aug 2026. (conference-board.org)
This is an early “yellow,” not a “red.” -
Atlanta Fed GDPNow: dashboard shows 1.8%, but the official Atlanta Fed update for Sep 25 shows 5.0% for 2026:Q3. (atlantafed.org)
For the score, the key is that nowcast is expansionary, not recessionary—even acknowledging model volatility.
Markets (risk-on, but stretched)
- Equity indices in the historical window trend up; the dashboard shows S&P 500 7743 / NASDAQ 27069 near highs.
This is a “no stress” read—but valuation extremes (NASDAQ/GDP, etc.) raise fragility to shocks, not necessarily recession by themselves.
Stock Screener Signals
Today’s quant flags cluster into two buckets: (1) value/dividend yielders (financials, telecom, selected international) and (2) oversold growth.
Bucket 1: “Value dividend” tilt (ARCC, AIG, FNF, T, BCE, HMC, LTM).
This looks like a market that still wants carry + cash-flow visibility—a common late-cycle posture when investors believe growth will slow but not collapse. Financials and credit-adjacent names (e.g., ARCC, AIG, FNF) also suggest the market is comfortable enough with default risk not to demand deep discounts—consistent with tight HY spreads.
Bucket 2: “Oversold growth” (CHTR, TLK).
These look like idiosyncratic drawdowns and mean-reversion setups rather than a broad “panic” screen. If recession risk were surging, you’d typically see the screen dominated by cyclicals in distress and a wider cross-section of defensives with strengthening momentum; instead, we see selective oversold flags inside an otherwise risk-tolerant tape.
One data-quality note you should treat as a red flag in the screener feed: several yields shown (e.g., ARCC 1002%) are not economically plausible as true dividend yields—more likely a parsing/annualization artifact. The category signal (value/carry tilt) is still useful, but don’t over-interpret the raw yield numbers.
Latest Economic Developments
-
Fed policy: The Federal Reserve raised the federal funds target range to 3.75%–4.00% on Sep 16, 2026. (federalreserve.gov)
In follow-up commentary, officials have emphasized concern that inflation may remain stuck above 2%, supporting the case for tighter policy longer. (apnews.com) -
Growth nowcast: Atlanta Fed GDPNow’s latest public update shows 2026:Q3 real GDP tracking at 5.0% as of Sep 25, 2026. (atlantafed.org)
That’s incompatible with “recession is already here,” and it explains why credit and equities remain relaxed. -
Sentiment: The University of Michigan’s final September 2026 sentiment index is 48.1, down from 51.7 previously reported in the table and described as the lowest reading in four months. (sca.isr.umich.edu)
This is the most recession-adjacent development because sentiment can become self-fulfilling if it changes spending behavior. -
Labor high-frequency: DOL data show initial claims of 197,000 for the week ending Sep 19, still consistent with low layoffs. (content.govdelivery.com)
-
Markets (tone check): Recent tape action remains broadly constructive—major indexes have been hovering near record territory in late September, reflecting risk-on positioning despite higher yields and geopolitical headlines. (apnews.com)
Near-Term Outlook (Next 30 Days)
Base case for the next month: MODERATE risk holds, with the score likely oscillating in the low-to-mid 30s unless one of two things happens:
-
Labor deterioration accelerates
- Watch weekly initial claims for a sustained move higher from the ~200K zone. (Today’s 197K is the opposite of that.) (content.govdelivery.com)
- Watch unemployment for a move above 4.1% and evidence of broadening (rising insured unemployment / slower quits).
-
Credit reprices (the fastest path to a higher score)
- HY OAS near 2.80% is extremely inconsistent with imminent recession. (ycharts.com)
- A meaningful widening (e.g., a fast move toward the mid-300s+) would change the model regime quickly.
Catalysts to monitor over the next 30 days:
- Ongoing Fed communications post-Sep meeting (guidance about whether September was “one-and-done” or the start of a mini-tightening sequence). (federalreserve.gov)
- Consumer-facing data (spending/income) to validate whether sentiment pessimism is translating into real restraint.
- Housing prints to see whether starts rebound is durable or noise.
Long-Term Outlook (3-6 Months)
The 3–6 month horizon is where today’s “scouts” matter most. The dashboard is essentially split:
-
Anti-recession forces (near-term support):
- Claims are low and falling. (content.govdelivery.com)
- Financial conditions are easy (NFCI -0.56). (fred.stlouisfed.org)
- Credit is relaxed (HY OAS ~2.80%). (ycharts.com)
- GDPNow is strong for Q3 (5.0%). (atlantafed.org)
-
Pro-recession setup (medium-term vulnerability):
- Fed just tightened into weak sentiment and soft goods-cycle signals. (federalreserve.gov)
- Temporary help and freight weakness are classic “early warnings” that can lead broad payrolls and production by months.
Historical parallel (pattern, not prophecy): softening in temporary help + freight alongside a still-healthy headline labor market is often the precondition for recession, not the recession itself. Recession risk rises meaningfully only if those early signals are followed by: (a) higher claims, (b) falling quits/job openings, (c) widening credit spreads, and (d) a sustained slide in leading indicators (LEI). Right now, only the first layer is convincingly flashing.
What to Watch
Hard thresholds / tripwires
- Initial claims: sustained break above ~230K–250K would be the first “regime change” signal (vs today’s 197K). (content.govdelivery.com)
- HY OAS: watch for a move from ~2.8% toward 3.5%+ as a sign credit is repricing growth risk. (ycharts.com)
- NFCI: a move toward 0 (tightening) would indicate conditions are no longer cushioning the cycle. (fred.stlouisfed.org)
High-signal releases
- Next reads on labor texture: quits, openings, weekly claims trend.
- Consumer spending/income prints to test whether sentiment 48.1 is “just vibes” or a real drag. (sca.isr.umich.edu)
- Conference Board LEI follow-through after the -0.1% August dip. (conference-board.org)
- Fed messaging: whether the Committee frames Sep 16 as a “risk-management hike” or the start of further tightening. (federalreserve.gov)