Recession Risk 34/100 — October 3, 2026
US recession risk over the next 90 days is MODERATE, not elevated, because the highest-weight real-time labor triggers are not flashing: the Sahm Rule remains untriggered and initial jobless claims just fell to 197,000 (week ending Sep 26, 2026). Payroll growth did, however, abruptly slow to +29,000 in September 2026 while unemployment held at 4.2%, a clear deceleration signal that bears watching closely. The yield curve has re-steepened (2Y ~4.83% vs 10Y ~5.28% on Oct 2, 2026), which is less recessionary than inversion but can reflect late-cycle dynamics. Forward indicators are mixed: the Conference Board LEI slipped -0.1% in August 2026, while manufacturing appears to be stabilizing/improving per the latest ISM commentary, but consumer sentiment remains extremely depressed (September 2026 down to ~48, with August at 51.7).
Recession Risk Score: 34/100 — MODERATE (+0 vs 30 days ago)
Today’s Recession Risk Score is 34/100 (MODERATE), unchanged vs 30 days ago. The macro picture still looks like a slowdown rather than an imminent contraction: the highest-weight real-time labor triggers are not flashing, with the Sahm Rule untriggered and initial jobless claims low. At the same time, the September payroll report delivered a sharp deceleration signal that keeps the score anchored in “moderate” rather than drifting lower. The market backdrop remains supportive (equities near highs, volatility contained), but several late-cycle fragilities (temporary help, freight, and “fear” ratios in metals) prevent a clean bill of health.
Score Trend — Last 30 Days
Over the last 30 days (window 2026-09-03 → 2026-10-03), the score started at 34 and ended at 34 (Δ 0), with a min of 34, max of 39, and average of 35 across 31 samples. The profile is best described as stable-with-spikes rather than trending: the baseline stayed pinned in the mid-30s, interrupted by a few brief jumps into the high-30s.
The last 10 readings show that shape clearly: the score printed 38 on 9/24, dropped back to 34 on 9/25, then stayed largely flat until another pop (38 on 9/30, 37 on 10/1) before reverting again to 34 on 10/2 and 10/3. That behavior—mean reversion after risk flare-ups—is consistent with a regime where recession risk is not yet self-reinforcing, but the system is sensitive to incremental labor and rates shocks.
Key Drivers
-
Labor stress remains contained (still the #1 anchor)
- Sahm Rule: 0.00 (SAFE) — no trigger.
- Initial jobless claims: 197K (SAFE) (week ending Sep 26, 2026) — consistent with a labor market that is cooling but not breaking.
- SOS recession indicator: 1.10 (SAFE) — insured unemployment not signaling broad stress.
-
Payroll growth abruptly decelerated (late-cycle warning)
- September 2026 payrolls: +29,000; unemployment: 4.2% (WATCH). (bls.gov)
- This is the cleanest “hard-data” slowdown impulse in today’s stack: even if the unemployment rate is still low, payroll momentum has become fragile enough that a second weak month would materially raise near-term recession odds.
-
Rates and curve dynamics: less inversion risk, more term-premium/late-cycle messaging
- 2s10s: +0.45 (WATCH), with market context showing 2Y ~4.83% vs 10Y ~5.28% on Oct 2, 2026—a positive curve (less classically recessionary than inversion). (dwkitalpha.com)
- However, the “why” matters: higher long rates can tighten financial conditions and weigh on housing/capex even if the curve is positive.
-
Financial conditions remain easy; credit isn’t screaming—yet
- Chicago Fed NFCI: -0.55 (SAFE) — loose conditions.
- HY OAS: ~324 bps (WATCH) — wider than the most benign regime, but not “stress” territory.
- This combination reads as: markets are not pricing a near-term recession, but are no longer pricing “perfect.”
-
Sentiment and goods-cycle signals are the recession-adjacent red flags
- UMich sentiment: crisis-level (DANGER) — extremely depressed vs typical expansion norms.
- Freight Transportation Index: -0.7 (DANGER) — goods economy weakening.
- Copper-to-gold ratio: extreme (DANGER) — “industrial fear” signal.
- These are not sufficient to “call” recession alone, but they help explain why the score won’t fall below the low-30s.
Category Breakdown
Using today’s signal counts:
-
Primary Indicators: 3 safe / 4 watch / 2 danger
Mixed, but the key takeaway is labor is not yet in a claims-driven deterioration, while select labor-leading components (temporary help, quits) are cautionary. -
Secondary Indicators: 2 safe / 0 watch / 1 danger
Mostly stable, but the one danger flag here matters because secondary families often turn only after the primary cycle has shifted. -
Housing & Construction: 0 safe / 1 watch / 1 danger
Housing remains a soft pocket: starts are weak and permits are cooling—consistent with a higher-rate drag. -
Business Activity: 2 safe / 1 watch / 0 danger
Business activity looks slowdown-but-not-contraction, aligning with the “moderate risk” score rather than “high.” -
Consumer Credit Stress: 1 safe / 3 watch / 0 danger
This is creeping stress, not a blow-up: delinquency and debt service are elevated enough to cap growth but not yet pointing to an abrupt collapse. -
Market Signals: 7 safe / 2 watch / 5 danger
A split tape: indexes and volatility look calm, while valuation/market-to-GDP style indicators and select macro ratios scream late-cycle excess. -
Liquidity: 0 safe / 1 watch / 2 danger
Liquidity is the fragile underbelly: the ON RRP depletion dynamic is a nontrivial “plumbing” change that can amplify rate volatility. -
Real-Time / High-Frequency: 0 safe / 1 watch / 1 danger
Real-time reads are not recessionary overall, but they’re brittle—especially if labor softening broadens.
Biggest Movers
Top 5 by absolute 7-day % change:
-
ON RRP Facility ($2B): +575.0% (7D)
Contradictory / mixed: the direction here is more about money-market plumbing than growth, but sharp moves can coincide with rate volatility and funding sensitivity. -
NY Fed Recession Probability (1.8%): -88.7% (7D)
Contradictory (improving): a steep drop is consistent with lower model-implied recession odds, supportive of the “slowdown, not recession” base case. -
Yield Curve (2s30s) (0.83): -15.8% (7D)
Confirmatory (worsening at the margin): curve moves can reflect shifting growth/inflation expectations; a fast flattening can be a caution flag even if still positive. -
VIX (16.4): +11.1% (7D)
Confirmatory (worsening): volatility rising from low levels often signals growing macro uncertainty, even if the absolute level remains benign. -
Initial Jobless Claims (197K): -10.1% (7D)
Contradictory (improving): a meaningful decline argues against the “imminent recession” narrative—claims are still behaving like a healthy market.
90-Day Indicator Trends
Important constraint: the provided “90-day history” block includes mostly observations clustered around early-to-late July 2026 for many series (not a full daily 90-day panel). Where the history is sparse, trend statements below are limited to what’s actually shown plus today’s stated reading.
Labor & income
- Initial claims: improved from 215K (Jul 5) → 208K (Jul 17) → 187K (Jul 25) in the history block, and today sits at 197K. Net: still low, with a mid-summer trough and a mild re-rise into late September that remains non-threatening.
- Unemployment rate: flat at 4.2% throughout July in the history block and 4.2% today—stable, but “WATCH” because the risk is asymmetry (a sustained rise would move the Sahm Rule quickly).
- Sahm Rule: 0.07 in July → 0.00 today (SAFE). Directionally, this is improving, reinforcing that the “Sahm-trigger” labor stress regime is not here.
Housing
- Building permits: drifted from 1410K (Jul 5) → 1367K (Jul 18) → 1374K (Jul 25); today is 1403K (WATCH). That reads like stabilization, not recovery.
- Housing starts: showed a jump in July history (1177K early July → 1427K mid/late July), but today is 1275K (WARNING)—suggesting housing is choppy and rate-sensitive, with downside risk if long rates stay elevated.
Financial conditions & credit
- NFCI: eased from around -0.50 to -0.55 by late July; today remains -0.55 (SAFE)—financial conditions are still loose.
- HY credit spreads: widened from ~267–275 bps early July to 277 bps late July in the history block; today shows 324 bps (WATCH), implying a material widening over the broader window. That’s not “panic,” but it is a real tightening impulse worth tracking.
Market/valuation excess (risk not recession—until it is)
- Equity indexes in the history block (July) were lower than today (e.g., S&P 500 ~7483–7575 in July history vs 7723 today), suggesting continued wealth-effect support.
- Yet the “danger” cluster (NASDAQ/GDP, copper-gold, valuation multiples) signals the market is pricing soft landing + duration, which can become fragile if earnings or rates disappoint.
Fiscal drag (slow burn)
- National debt and interest expense in the history block are already high (e.g., interest expense ~$1219B in July history vs ~$1280B today), implying worsening fiscal carrying costs. This is not a 90-day recession trigger, but it’s a medium-term constraint that can amplify future downturns.
Stock Screener Signals
Today’s quant screen is dominated by “value dividend” flags (ARCC, AIG, BBY, FNF, HMC, T, BCE, LTM) and a smaller “oversold growth” pocket (CHTR, TLK). The macro interpretation: the market is still willing to own risk (indexes near highs), but positioning appears to be tilting toward cash-flow yield and lower multiples—a classic late-cycle preference when investors want equity exposure without paying peak multiples.
Two cautions jump out:
- The screen’s “yield” figures are implausibly high for several tickers (likely a data normalization issue), so treat yields as a style label rather than a literal payout signal.
- CHTR (RSI 28) and TLK (RSI 30) suggest a mean-reversion bid in select growth/EM telecom—often consistent with risk-on pockets inside a broader “quality/value” rotation, not a recession panic.
Net: the screener reads as defensive-but-not-fearful—consistent with a 34/100 risk score.
Latest Economic Developments
-
September 2026 jobs report (released Oct 2, 2026): Payroll growth slowed sharply to +29,000 and unemployment ticked up to 4.2%. (bls.gov)
This is the single most important “fresh” macro development for recession risk. It doesn’t confirm recession by itself, but it raises the probability that labor demand is transitioning from cooling to weakening, especially if revisions are negative or if temporary help keeps contracting. -
Treasury market stress / high long yields: Market commentary over the past 48 hours focused on spiking Treasury yields and the possibility of a structurally higher-rate world, with the 10-year around the mid-5% area after touching levels last seen decades ago. (axios.com)
The recession channel here is indirect: higher term yields tighten mortgage rates, corporate discount rates, and fiscal interest costs—even if the curve is now positive. -
FOMC: Sep 16, 2026 rate hike and guidance backdrop: The Fed raised the target range by 25 bps to 3.75%–4.00% at the Sep 16, 2026 meeting. (federalreserve.gov)
That decision matters now because the economy is presenting late-cycle mixed signals: weak payroll growth plus high long rates can do some of the Fed’s tightening “for it,” which increases the probability of a pause—unless inflation re-accelerates.
Near-Term Outlook (Next 30 Days)
Base case for October: growth slowdown, recession risk range-bound unless labor cracks.
Key catalysts likely to move the score within the next month:
- October employment report (BLS): scheduled for Friday, November 6, 2026. (bls.gov)
A second consecutive weak payroll print (or a clear rise in unemployment) would push multiple labor components from WATCH → WARNING/DANGER quickly. - Weekly claims: watch not just the level (still excellent) but the trend—a sustained move above the low-200Ks would be the earliest “real-time” confirmation that layoffs are spreading.
- Credit spreads: HY OAS is the key market-based tripwire; a move from ~324 bps toward materially wider levels would be confirmatory of rising default/slowdown risk.
- Rates volatility: if long yields remain elevated, housing and interest-sensitive consumption weaken further, raising the likelihood that “soft landing” becomes “stall speed.”
Long-Term Outlook (3-6 Months)
Over the next 3–6 months, recession risk depends on whether the economy can absorb a higher-for-longer term structure without turning a hiring slowdown into a layoff cycle.
Three structural themes dominate:
- Labor is the fulcrum. Claims and the Sahm Rule say “no recession,” but payroll momentum has slowed to a near-stall. If unemployment rises persistently from 4.2% and the Sahm Rule begins climbing, recession odds would increase rapidly because labor weakness tends to be nonlinear once it starts.
- The yield curve is no longer the headline risk—term premium is. A positive 2s10s reduces the classic inversion signal, but high long yields can still tighten conditions enough to trigger a downturn with a lag.
- Market calm can coexist with macro fragility. With equities near highs and VIX low, financial conditions remain supportive—yet several “danger” indicators (temporary help, freight, copper/gold, valuation-to-GDP measures) suggest the system is late-cycle and vulnerable to shocks.
Net: if labor stabilizes (claims stay low; payrolls rebound modestly), risk likely drifts sideways to slightly lower. If labor deteriorates (claims trend up; unemployment rises), risk can re-rate higher quickly even without a credit event.
What to Watch
Hard thresholds and near-term triggers:
- Initial jobless claims: watch for a sustained move above ~230K–250K (trend, not one-week noise).
- Unemployment rate: any persistent rise above 4.2% (and especially another 0.2–0.3 pp up move over a few months) would pressure the Sahm Rule.
- Temporary help services: already DANGER—watch whether this weakness spreads into broader payroll categories.
- HY OAS: watch for widening beyond the current ~324 bps into a clearly risk-off regime.
- Long-end yields: if the 10-year remains pinned near mid-5% or rises again, expect housing and rate-sensitive sectors to weaken further.
- Next payroll release date: November 6, 2026 (October employment report). (bls.gov)
Sources
No data available for this window.