Recession Risk 34/100 — October 5, 2026

Recession Risk Score: 34/100 — MODERATE (+0 vs 30 days ago)

Today’s Recession Risk Score holds at 34/100 (MODERATE), unchanged versus 30 days ago. The macro message is “slowdown risk, not recession start”: high-frequency labor stress remains contained (claims are still sub-200k), and the curve is re-steepened, which typically reduces immediate recession odds relative to an inversion regime. The tension is that the September payroll report was very soft (+29k; unemployment 4.2%), while sentiment is deeply depressed, leaving the expansion vulnerable to a confidence-to-spending downdraft. Net: conditions warrant vigilance, but the highest-weight “recession has begun” triggers still haven’t confirmed.

Score Trend — Last 30 Days

Over the last 30 days (2026-09-05 → 2026-10-05), the score started at 34, ended at 34, and printed a min of 34 / max of 39 (average 35, 31 samples). That’s a classic range-bound “moderate risk” regime: the system is seeing intermittent stress signals, but not enough breadth or persistence to push into an elevated band.

The shape matters: the late-window bump to 38–37 on Sep 30–Oct 1 looks like a temporary flare-up rather than a trend break, with the score snapping back to 34 by Oct 2 and staying there through Oct 5. In plain English: the macro tape is mean-reverting—soft patches show up, but they’re not yet propagating through labor, credit, and financial conditions at the same time.

Key Drivers

1. Labor “hard” data cooled, but claims haven’t cracked

   • The September employment report showed nonfarm payrolls +29,000 and unemployment at 4.2%. (bls.gov)
   • Yet initial jobless claims remain very low: 197,000 (recent weekly print), signaling layoffs are not broad-based. (apnews.com)
   • Translation: hiring is slow, but separation pressure is still benign—this delays the usual recession cascade (income → consumption → layoffs).

2. Sahm Rule remains “safe,” reinforcing “not yet”

   • Your tracker shows Sahm Rule = 0.00 (SAFE), meaning the unemployment-rate acceleration signal that often tags recession onset is not close to firing.
   • This is a key anchor holding the score in the 30s despite the weak payroll headline.

3. Yield curve has re-steepened (less immediate recession pressure)

   • 2s10s is ~+45 bps (WATCH), consistent with “post-inversion steepening.” A positive curve usually implies near-term recession odds are lower than during inversion, even if it’s not an all-clear.
   • 2s30s is +0.83 (SAFE), reinforcing the “not inverted” message.

4. Forward-growth and broad leading signals are not collapsing

   • The Conference Board LEI edged down just -0.1% m/m in August 2026 (after +0.2% in July), with the level at 99.5—not the kind of persistent deterioration typically seen before sharp contractions. (conference-board.org)
   • NY Fed Staff Nowcast for 2026:Q3 is ~2.3% (with a wide probability interval), pointing to ongoing expansion in the near-term tracking data. (newyorkfed.org)

5. Financial conditions are loose, cushioning the slowdown

   • Chicago Fed NFCI sits around -0.55, indicating looser-than-average conditions (supportive for risk assets and credit availability). (fred.stlouisfed.org)
   • This helps explain why equities can sit near highs even as survey psychology looks awful.

6. Sentiment is recessionary even if activity isn’t

   • University of Michigan sentiment final = 48.1 for September 2026 (today’s reading in your dashboard is 51.7, still “danger” territory). (sca.isr.umich.edu)
   • Ultra-low sentiment is a downside risk amplifier: it can become “real” if it restrains discretionary spending and big-ticket purchases—especially alongside rising consumer credit stress signals.

Category Breakdown

Using the provided signal counts:

• Primary Indicators: 3 safe / 4 watch / 2 danger
  Mixed but not recession-confirming: the center of gravity is still “watch,” anchored by low claims and a safe Sahm reading, offset by labor-leading weak spots.

• Secondary Indicators: 2 safe / 0 watch / 1 danger
  Secondary is mostly fine, but the single danger reading matters because second-tier cracks often precede broader labor deterioration.

• Housing & Construction: 0 safe / 1 watch / 1 danger
  Housing remains a pressure point (permits/starts not rebounding convincingly), consistent with rate sensitivity and affordability drag.

• Business Activity: 2 safe / 1 watch / 0 danger
  Business activity is still net supportive, helped by an expanding manufacturing PMI regime.

• Consumer Credit Stress: 1 safe / 3 watch / 0 danger
  No acute blow-up, but the “watch” cluster (delinquencies, debt service, savings) signals reduced resilience if labor weakens further.

• Market Signals: 7 safe / 2 watch / 5 danger
  Markets are internally bifurcated: indices and volatility look fine, while valuation/ratio-style indicators (and some macro market ratios) flash risk.

• Liquidity: 0 safe / 1 watch / 2 danger
  Liquidity looks late-cycle fragile: RRP depletion and other balance-sheet/liquidity proxies can matter quickly if volatility returns.

• Real-Time / High-Frequency: 0 safe / 1 watch / 1 danger
  The real-time bucket isn’t broadly recessionary, but it is sensitive: one or two labor prints can swing this category materially.

Biggest Movers

Top 5 by |7-day % change|:

1. ON RRP Facility ($2B): +1026.0% (7D)
   Mechanically huge % move off a tiny base; directionally, a rising print from near-depletion is not automatically recessionary, but the broader “RRP drained” backdrop is a liquidity regime change (watch for knock-on effects in funding markets).

2. NY Fed Recession Probability (1.8%): +742.9% (7D)
   Big percentage move, but the level is still low (1.8%), so this is more a contradictory/false-alarm style uptick than confirmation—unless it continues rising.

3. Yield Curve (2s10s) (0.45): +30.6% (7D)
   A steeper curve is generally contradictory (improving) for near-term recession risk versus inversion conditions.

4. Yield Curve (2s30s) (0.83): -12.9% (7D)
   Still positive; the decline is mild and not confirmatory unless it trends toward flattening/inversion.

5. Consumer Sentiment (UMich) (51.7): +10.5% (7D)
   Improvement is contradictory (improving), but the level remains danger—this is “less bad,” not “good.”

90-Day Indicator Trends

The 90-day history provided covers primarily early July → early August 2026 for many series, so trend read-through is about direction of travel into the late-summer baseline, then cross-referenced with today’s snapshot.

Labor & income (high weight)

• Initial claims: moved from ~215k (Jul 7) → 187k (Jul 25) → 197k (Aug 1), and today remains 197k (SAFE). That’s a net improvement vs early July and remains consistent with “no layoffs wave.”
• Unemployment rate: held at 4.2% across the July–Aug window and is 4.2% today (WATCH). Stability here is why the Sahm Rule stays safe (your history shows Sahm ~0.07 in July/Aug; today’s dashboard shows 0.00).
• Real personal income (ex transfers): $16.6T in July/Aug window (WATCH), while today reads $17.0T (WATCH)—a modest upshift that supports consumption if confidence doesn’t choke it off.

Financial conditions & credit

• NFCI: eased from about -0.50 (Jul 7) to around -0.55 (Aug 1) and sits near -0.55 today—a steady “loose conditions” backdrop. (fred.stlouisfed.org)
• HY OAS (credit spreads): in the history, ~274 bps (Jul 7) rising to ~281 bps (Jul 29); today is 324 bps (WATCH). That’s a meaningful widening vs early summer, not crisis, but consistent with “late-cycle repricing” that can become self-reinforcing if earnings or defaults rise.

Housing & construction

• Housing starts: jumped in the history from 1177k (Jul 7) to 1427k (Jul 18–Aug 1), but today is 1275k (WARNING)—a reversal lower versus late-July strength.
• Building permits: slipped from 1410k (Jul 7) to ~1374k (Aug 1); today is 1403k (WATCH)—slightly better than early August but still not a decisive uptrend.
  Net: housing remains a soft underbelly, consistent with your category score.

Production, manufacturing, and goods flow

• Industrial production index: ~102.6–102.64 in the July–Aug window; today is 103.1 (SAFE)—a mild improvement.
• Manufacturing PMI: September print 54.5 (expansion), essentially flat vs August (54.6); the Employment Index improved to 52.7. (ismworld.org)
• Freight transportation index: stuck at 0.3 (danger) through the window; today shows -0.7 (DANGER), implying the goods-moving economy is weakening even while PMI reads strong—an internal contradiction worth respecting.

Household resilience

• Personal savings rate: deteriorated from 3.0% in July to 2.7% by Aug 1 (danger in your framework); today reads 4.1% (WATCH)—better than the early-Aug trough, but still not high enough to be a robust buffer.
• Credit card delinquency: essentially flat around 2.9% through July/Aug; today remains 2.9% (WATCH)—elevated but not accelerating in the provided history.

Stock Screener Signals

Today’s quant flags are dominated by “value dividend” screens: ARCC, AIG, BBY, FNF, HMC, T, LTM, BCE—plus a smaller “oversold growth” pocket (CHTR, TLK). The macro interpretation is straightforward: the market is selectively seeking yield and cash-flow durability, while still hunting mean-reversion in names that have been hit hard enough to register oversold momentum.

Two things stand out:

• Defensive-ish cash flow and balance-sheet preference: Finance/insurance (AIG), telecom (T, BCE), and asset-based yield vehicles (ARCC) show up when investors want carry and valuation support if growth momentum wobbles. This aligns with a moderate recession-risk regime—positioning for slower growth, not an imminent collapse.
• Oversold growth signals are idiosyncratic rather than broad risk-off: CHTR (RSI 28) and TLK (RSI 30) look like targeted drawdowns, not panic. In a true “recession-now” tape you’d often see broader cyclicals break simultaneously and defensives dominate exclusively.

One caution: several listed dividend yields are obviously data artifacts/outliers (e.g., triple-digit yields). Treat the style signal (value/dividend tilt) as informative, not the raw yield magnitudes.

Latest Economic Developments

1) The labor market just delivered a “growth scare” payroll print.
The Bureau of Labor Statistics reported +29,000 payrolls in September 2026 and unemployment at 4.2%. (bls.gov) This is not recession confirmation by itself, but it is exactly the kind of downside surprise that can tighten financial conditions if repeated (via weaker earnings expectations and wider credit spreads).

2) Weekly claims remain exceptionally calm—still the best real-time recession antidote.
Initial jobless claims recently printed 197,000, the lowest since mid-July per reporting, and consistent with low layoffs. (apnews.com) In near-term recession calls, claims are one of the fastest “truth meters.” Today, that meter says no layoffs wave.

3) The Fed tightened again in mid-September, keeping policy risk on the table.
The Federal Reserve raised the target range to 3.75%–4.00% on September 16, 2026. (federalreserve.gov) This matters for recession odds not because one hike breaks the economy, but because late-cycle hikes raise the probability of a policy-induced air pocket if labor cooling accelerates.

4) Leading indicators softened marginally, but not in a “downshift spiral.”
The Conference Board LEI fell -0.1% m/m in August 2026 to 99.5 after rising in July. (conference-board.org) That’s a yellow flag, but not a collapse. Meanwhile, the NY Fed Staff Nowcast for 2026:Q3 is ~2.3% as of the Sep 18 data flow, consistent with continued expansion in tracking estimates. (newyorkfed.org)

5) Manufacturing remains a support beam (for now).
ISM Manufacturing PMI is 54.5 in September, with the Employment Index at 52.7—both expansionary readings. (ismworld.org) If recession risk rises later this quarter, it will likely be because labor/consumption weaken—not because manufacturing is already in contraction.

Near-Term Outlook (Next 30 Days)

Base case for October 2026: slowdown scare with elevated headline volatility, but no recession trigger confirmation unless labor data deteriorate rapidly.

Key catalysts over the next month:

• Weekly initial jobless claims: the market will react quickly to a regime change. Your stated threshold zone (~230k–250k) is sensible as a “something changed” level; staying near ~200k keeps the recession score anchored.
• Next jobs report timing: The BLS notes the October 2026 Employment Situation is scheduled for Friday, November 6, 2026—a major event risk for the score trajectory beyond the next 30 days, but positioning will begin well before then. (bls.gov)
• Nowcast/LEI updates: Any downgrade cycle in tracking estimates (NY Fed nowcast sliding toward ~1% or lower) combined with a second weak payroll could push the score into the 40s quickly.
• Credit spreads: HY OAS at ~324 bps is “watch,” not distress. But widening toward ~400+ bps would be a meaningful confirmation that markets are pricing a sharper growth hit.

Long-Term Outlook (3-6 Months)

The 3–6 month horizon is about whether the economy transitions from cooling to contraction—and that decision tree runs through labor and credit.

• Constructive path (score drifts down toward low-30s/high-20s):
  Claims stay sub-220k, unemployment stabilizes near ~4.2%, and the curve remains positively sloped. In that world, depressed sentiment is a headwind but not decisive, and manufacturing expansion helps prevent a broader rollover.

• Adverse path (score rises into the 40s–50s):
  Payroll softness persists (multiple sub-50k prints), unemployment edges higher, temp help continues to contract (your dashboard already flags Temporary Help Services as DANGER), and HY spreads widen. That combination would look like a classic late-cycle labor turning point: hiring freezes first, then layoffs follow with a lag.

The key macro asymmetry is that today’s conditions are “stable until they aren’t.” When claims turn, they can turn fast. The fact that policy has recently tightened (Sep 16 hike) increases the chance that a small negative shock becomes a bigger one through confidence and credit channels. (federalreserve.gov)

What to Watch

Labor (highest weight, fastest confirmation)

• Initial claims: sustained move above ~230k–250k (your watch zone) would be an early warning.
• Unemployment rate: if the 3-month average rises materially and your Sahm tracker starts climbing off 0.00, risk will reprice quickly.

Credit & liquidity (propagation mechanism)

• HY OAS: watch 324 bps → 400+ bps as a regime change threshold.
• Liquidity plumbing: with RRP near depletion, monitor signs of funding stress or abrupt money-market rate volatility (even if the RRP level itself is noisy day-to-day).

Housing

• Starts/permits: stabilization matters; renewed declines would reinforce the rate-sensitivity drag and raise risk of broader cyclical spillover.

Growth tracking

• NY Fed Staff Nowcast: watch for a trend down from ~2.3% toward ~1% or lower as data flow updates. (newyorkfed.org)
• Conference Board LEI: a single -0.1% print is manageable; a sequence of negative prints is not. (conference-board.org)

Sources

• apnews.com
• axios.com
• axios.com
• bls.gov
• conference-board.org
• federalreserve.gov
• federalreserve.gov
• fred.stlouisfed.org
• govdelivery.com
• ismworld.org
• kiplinger.com
• kiplinger.com
• newyorkfed.org
• sca.isr.umich.edu