Recession Risk 38/100 — September 30, 2026

Recession Risk Score: 38/100 — MODERATE (+4 vs 30 days ago)

Today’s Recession Risk Score is 38/100 (MODERATE), up +4 points from 30 days ago (34 → 38). The increase reflects a widening gap between “soft” recessionary signals (sentiment, freight/goods, select labor leading edges) and still-stable hard labor data (claims, Sahm Rule). The macro picture remains slow-and-brittle rather than contractionary, but the risk distribution is getting more asymmetric: if labor weakens, the downside could accelerate quickly.

Score Trend — Last 30 Days

Over the last 30 days (2026-08-31 → 2026-09-30), the score drifted higher from 34 to 38 (+4). The range was tight—min 34, max 39, average 35—which tells you this is not a regime change into “high risk,” but rather a persistent, moderate-risk grind with occasional stress spikes.

The shape matters. The series spent much of the last 10 days pinned at 34, then jumped to 38 today after a short burst earlier in the window (38–39 on Sep 22–24). That pattern is consistent with mean-reversion around the mid-30s punctuated by event-driven jolts (typically policy and labor-market headlines). Net: recession risk is not trending relentlessly higher, but it is failing to reset lower, which is what you’d want to see in a clean soft-landing glidepath.

Key Drivers

1. Labor “trigger” remains untriggered, but leading edges are fraying

   • Sahm Rule: -0.07 (SAFE) — still clearly below trigger; no broad-based unemployment acceleration signal yet.
   • Initial jobless claims: 197k (SAFE) in the Sep 24, 2026 release—historically low and consistent with limited layoffs. (apnews.com)
   • JOLTS (Aug 2026): openings ~7.1M; quits rate 1.9% — lower dynamism and worker confidence. (bls.gov)
   • Temporary Help Services: 2,520k (DANGER) — this is the loudest labor-leading warning in your dashboard; temp staffing usually rolls over before payrolls.

2. Policy risk: restrictive tilt reinforced by the September hike

   • The Fed raised rates by 25 bp on Sep 16, 2026 to 3.75%–4.00%. (investing.com)
   • In a slow-growth environment, additional tightening raises the probability of a policy-induced stall (especially if labor demand cools further and credit-sensitive consumers crack).

3. Financial conditions remain loose (a cushion), not tight (a trigger)

   • Chicago Fed NFCI: -0.56 (SAFE) — consistent with loose conditions, not systemic stress. (fred.stlouisfed.org)
   • HY OAS: ~270 bps (SAFE) — still tight, signaling limited near-term default stress in markets (for now).

4. Sentiment is recessionary—even if spending hasn’t fully broken

   • UMich sentiment: 51.7 (DANGER) (August reading), with September reportedly weaker in the UMich tracker. (sca.isr.umich.edu)
   • This “sentiment collapse without layoffs” setup often persists longer than expected—but when labor weakens, confidence tends to amplify the downturn.

5. Goods/transport weakness is persistent and consistent with a downshift

   • Freight Transportation Index: -0.3 (DANGER) — aligns with a soft goods economy and cautious inventories, often an early-cycle warning even when services hold up.

Category Breakdown

Using today’s signal counts:

• Primary Indicators: 3 safe / 4 watch / 2 danger
  Mixed. “Hard” recession triggers are mostly not firing, but the watch/danger mix says the labor-and-growth core is degrading at the margin.

• Secondary Indicators: 2 safe / 0 watch / 1 danger
  Narrow stress rather than broad confirmation.

• Housing & Construction: 0 safe / 1 watch / 1 danger
  Housing remains a growth drag; it’s not collapsing, but it’s not a tailwind either.

• Business Activity: 2 safe / 1 watch / 0 danger
  Business activity is soft but not breaking, consistent with “slow expansion.”

• Consumer Credit Stress: 1 safe / 2 watch / 1 danger
  This is a key convexity bucket: stress can rise quickly if labor softens, especially with low savings.

• Market Signals: 6 safe / 3 watch / 5 danger
  Markets are giving a conflicted message: index-level strength and low vol coexist with valuation/relative-price warnings (and defensive ratios like gold/silver).

• Liquidity: 0 safe / 1 watch / 2 danger
  Liquidity is less forgiving than earlier cycles; plumbing isn’t broken, but buffers look thinner.

• Real-Time / High-Frequency: 0 safe / 1 watch / 1 danger
  High-frequency is not yet recession-confirmatory, but it’s no longer clean.

Biggest Movers

Top 5 by absolute 7-day % move:

1. ON RRP Facility ($5B): -52.7% (7D)
   Confirmatory (worsening fragility). A near-depleted RRP can be consistent with less parked cash and potentially tighter marginal liquidity conditions (context-dependent, but directionally reduces “excess cushion”).

2. Housing Starts (1275K): +21.2% (7D)
   Contradictory (improving). Starts bouncing helps stabilize cyclicals; one week isn’t a trend, but it’s the right direction for “no recession.”

3. VIX (14.2): +13.4% (7D)
   Mildly confirmatory (worsening risk). Vol rising from very low levels often signals more two-way macro uncertainty, even if absolute VIX remains subdued.

4. Chicago Fed NFCI (-0.56): -7.2% (7D)
   Contradictory (improving). More negative NFCI = looser conditions, which reduces recession odds near term.

5. NY Fed Recession Probability (0.9%): +6.8% (7D)
   Confirmatory (worsening), but the level remains low enough that it’s more “noise around a low base” than a true alarm.

90-Day Indicator Trends

Your 90-day history window (as provided) is heavily concentrated around early July snapshots, but it still reveals direction-of-travel and key inflection behavior:

• Industrial Production: 102.6 → 103.1 (SAFE)
  The level is up modestly versus early July readings (102.6). This argues against an imminent hard downturn in “factory output,” even if freight is weak.

• Sahm Rule: 0.10 (Jul 2) → 0.07 (early July) → -0.07 (today)
  The labor-trigger picture is improving, not deteriorating—one of the strongest “no recession in the next ~90 days” supports in your model.

• Initial Claims: 215k (early July) → 208k (mid-July) → 197k (Sep 24 release)
  Claims trend has moved lower, consistent with continued labor-market resilience. (apnews.com)

• Yield Curve (2s10s): ~0.31–0.42 in early July; 0.37 today
  The curve is positive and broadly stable-to-slightly-steeper versus early July; that reduces the classic inversion warning. The analytical key is why it steepens: bull steepening (front-end falling on growth fears) would be less comforting than bear steepening (long end rising on term premium/inflation risk).

• Financial Conditions (NFCI): about -0.50 → -0.55 range
  Conditions are consistently loose, which tends to postpone recession even when sentiment deteriorates. (fred.stlouisfed.org)

• Consumer Sentiment: 44.8 (early July in your history) → 51.7 (today’s stated reading)
  Sentiment improved from extremely depressed levels but remains very weak in absolute terms—and UMich’s tracker shows renewed slippage in September. (sca.isr.umich.edu)

• Housing (permits/starts): permits slipped (1410k → 1367k in July history), starts show a later rebound in your history (1177k → 1427k) but today’s dashboard is weaker (starts 1275k; permits 1403k). Net: choppy housing, consistent with “rate-sensitive drag, not a 2008-style collapse.”

• Credit spreads: HY OAS stayed tight around the high-260s/low-270s (e.g., 267–275 bps in July history; 270 bps today). That’s inconsistent with imminent recession pricing.

Bottom line from the “90-day” lens: hard activity + hard labor look stable, while soft data + labor leading indicators (temp help, quits, openings) are the fault line. That is exactly the setup for a moderate (not high) recession score: recession risk is present, but it needs confirmation from payrolls/claims/unemployment.

Stock Screener Signals

Today’s screener is dominated by “value dividend” flags—ARCC, AIG, BBY, FNF, HMC, T, BCE—plus a couple oversold growth names (CHTR, TLK) with low RSIs (notably CHTR RSI ~28). That combination reads like barbell positioning: investors are still willing to own equities (even growth selectively), but they increasingly want cash-flow, dividends, and valuation support rather than pure cyclicality.

A notable red flag is the implausibly high dividend yields shown on several names (e.g., “1002%” for ARCC). Treat these as data-quality artifacts (often caused by special dividends, stale prices, split adjustments, or incorrect annualization). The real macro signal here is the style tilt: your quant lens is finding more opportunities in cheap, cash-generative balance-sheet stories than in high-multiple cyclicals—consistent with an economy where growth is positive but less reliable.

If this were a “risk-on, accelerating growth” environment, you’d expect more broad-based momentum/growth screens and fewer deep-value dividend clusters. Instead, this looks like a market that expects slower nominal growth and more dispersion—a regime where defensives and idiosyncratic bargains outperform, while high-beta cyclicals need clean macro upside to work.

Latest Economic Developments

• Labor market (latest JOLTS): The BLS reported job openings little changed at ~7.1 million in August 2026 and the quits rate unchanged at 1.9%. (bls.gov)
  Interpretation: this isn’t “recession labor,” but it is a sign the labor market is less dynamic—fewer quits and fewer openings tends to foreshadow slower wage pressure and softer consumption growth.

• Weekly claims: Initial unemployment claims fell to 197,000 (reported Sep 24, 2026), the lowest since mid-July per coverage—layoffs remain rare. (apnews.com)
  Interpretation: the recession call cannot be made from hard labor data today. Claims are one of the highest-signal short leading indicators; right now they argue “no imminent break.”

• Fed policy: Coverage and documentation confirm the Fed hiked 25 bp on Sep 16, 2026 to 3.75%–4.00%. (investing.com)
  Interpretation: policy risk is now skewed toward overtightening if growth slows further, particularly because several private/market indicators already show late-cycle stress in consumers (low savings, rising delinquencies in your dashboard).

• Consumer sentiment: UMich sentiment shows 51.7 in August and a weaker September reading (~48.1) in the UMich tracker. (fred.stlouisfed.org)
  Interpretation: households are behaving like recession is near even though the labor market hasn’t validated that fear—this mismatch is a classic brittleness signal.

Near-Term Outlook (Next 30 Days)

The next month is about labor confirmation vs. false alarm.

Base case (most likely): the score oscillates in the mid-to-high 30s as hard data stays “okay” but leading indicators stay “not okay.” Specifically:

• Claims probably remain sub-220k unless layoffs broaden. A move toward 230k–250k on a sustained basis would likely lift the risk score quickly.
• The September Employment Situation report (early October) is the key catalyst. A downside surprise in payrolls plus a rise in unemployment would push the model toward a higher band, because it would begin converting “leading-edge weakness” into “hard confirmation.”
• Markets remain vulnerable to a repricing if the Fed signals additional hikes (or if long rates push tighter financial conditions), but as long as spreads remain tight, equity weakness alone won’t confirm recession.

Long-Term Outlook (3-6 Months)

Over 3–6 months, the economy looks set up for continued deceleration with elevated tail risk, not a guaranteed contraction.

The reason: the labor market is the final domino, and right now it’s still standing (claims low; Sahm safe). But the supports under the domino are weakening:

• Temp help is already in DANGER (classic pre-payroll warning).
• Openings and quits imply slower job switching and softer bargaining power—often a precursor to a rising unemployment rate.
• Household buffers (low savings, rising delinquencies in your dashboard) increase the odds that consumption weakens nonlinearly if unemployment rises even modestly.

A useful historical parallel is the common “soft data cracks first” late-cycle phase: sentiment and goods soften, services hold up, and then labor either stabilizes (soft landing) or rolls over (recession). Your dashboard today still aligns more with soft landing risk than recession inevitability—but the longer temp help and freight remain weak, the more likely the labor market eventually follows.

What to Watch

Hard labor triggers

• Initial claims: sustained move above 230k, then 250k.
• Unemployment rate: sustained rise beyond 4.3%–4.5% would raise Sahm proximity and push the score higher.
• Sahm Rule: watch for a move toward 0.2–0.3; today’s -0.07 is safely away from trigger.

Leading labor + activity

• Temporary help: any further step-down from 2,520k would reinforce “payroll risk ahead.”
• JOLTS: openings below 7.0M and quits below 1.8% would strengthen the “cooling labor demand” narrative.

Credit and markets

• HY OAS: a break wider above 350 bps would be an early recession-pricing tell; staying near ~270 bps keeps the “no stress” baseline intact.
• Financial conditions (NFCI): a move toward zero (tightening) would be a material warning.

Policy

• Fed communication following the Sep 16, 2026 hike: any guidance that signals more tightening despite slowing indicators would raise overtightening odds.

Sources

No data available for this window.