Recession Risk 38/100 — September 10, 2026

Recession Risk Score: 38/100 — MODERATE (+0 vs 30 days ago)

Today’s Recession Risk Score is 38/100 (MODERATE), unchanged versus 30 days ago. The topline is stable because real-time labor stress remains muted (claims and the Sahm-style unemployment acceleration signal are not flashing), while several leading-cycle and market-valuation warnings continue to stack up. In short: the economy doesn’t look like it’s rolling over in the next 4–8 weeks, but it is becoming more brittle—meaning a confidence, energy, or policy shock would travel faster through the system than it would with healthier household buffers and cleaner financial plumbing.

---

Score Trend — Last 30 Days

Over the last 30 days (2026-08-11 → 2026-09-10), the score started at 38 and ended at 38 with no net change, but the path matters: the window saw a floor at 34 and a ceiling at 38, with an average of 36 across 31 samples. That’s a classic range-bound, mean-reverting profile rather than a persistent climb into “high-risk” territory.

The last 10 readings show a “sawtooth” pattern (34 ↔ 38) rather than a smooth drift: the score sat at 34 from Sep 1–3, jumped to 38 on Sep 4, flipped back to 34 on Sep 5, returned to 38 on Sep 6, then mostly held 34 until Sep 10, when it moved back to 38. This behavior is consistent with a macro backdrop where one cluster of indicators (labor + financial conditions) keeps pulling risk down, while another cluster (cyclical leading indicators + valuations + fiscal/liquidity fragilities) keeps pulling it back up.

Bottom line on the shape: the system is not accelerating toward recession, but it is not healing—it’s stabilizing at a moderate-risk plateau.

---

Key Drivers

1. Labor-market recession triggers remain “off” (biggest stabilizer)

   • Sahm Rule: -0.07 (SAFE) — far from the 0.50 recession trigger in common Sahm formulations.
   • Initial jobless claims: 206K (SAFE) — still consistent with limited layoff pressure; recent reporting notes claims remain historically low even as they ticked up. (apnews.com)

2. Payrolls rebounded in August, keeping “imminent recession” odds contained

   • August 2026 nonfarm payrolls: +162K
   • Unemployment rate: 4.1%
   • BLS confirmation anchors this as a solid labor print and reduces near-term recession probability. (bls.gov)

3. Nowcast growth is supportive—but the narrative is fragile

   • The Atlanta Fed’s GDPNow for 2026:Q3 was ~4.7% SAAR as of Sep 3, 2026, a strong estimate for late-cycle conditions. (atlantafed.org)
   • This reduces the probability that the economy is already in a contraction today, even if forward-looking indicators are deteriorating.

4. Late-cycle yield-curve steepening is a “watch,” not a signal—yet

   • 2s10s: +40 bps (WATCH) after inversion. Post-inversion steepening can coincide with late-cycle dynamics (growth slowing + easing expectations), but by itself it’s not a 90-day recession call without labor deterioration.

5. Leading-cycle deterioration is increasingly concentrated in “canaries”

   • Temporary help services: 2,520K (DANGER) — temp employment often rolls over early as firms reduce variable staffing.
   • Freight Transportation Index: -0.3 (DANGER) — goods/economic throughput is weakening.
   • Consumer sentiment: 55.2 (WARNING) — weak confidence conditions can transmit quickly into discretionary spending.

6. Tail risks are elevated by fiscal + financial-system convexity

   • Total U.S. national debt: $39.1T (DANGER)
   • Interest expense: ~$1.247T/yr (WARNING)
   • Bank unrealized losses: ~$5.155T (WARNING)
   • These don’t force a recession on a 90-day horizon, but they raise the odds of a fast downside break if funding conditions tighten.

---

Category Breakdown

• Primary Indicators: 3 safe / 4 watch / 2 danger
  Labor remains supportive, but leading labor components (notably temp help) are pulling the cycle risk higher.

• Secondary Indicators: 2 safe / 0 watch / 1 danger
  The mix is still supportive overall, but the “danger” component suggests early-cycle softening isn’t isolated.

• Housing & Construction: 0 safe / 1 watch / 1 danger
  Housing remains a key weak spot: starts are warning and permits are watch, consistent with rate sensitivity and affordability drag.

• Business Activity: 2 safe / 1 watch / 0 danger
  Business-side indicators look stable enough to prevent a near-term recession call, but “watch” signals imply momentum is not improving.

• Consumer Credit Stress: 1 safe / 2 watch / 1 danger
  Credit stress is not acute, but it’s creeping—the low savings cushion makes any labor softening more dangerous.

• Market Signals: 6 safe / 3 watch / 5 danger
  Markets are near highs and volatility is complacent, but valuation/fundamentals divergences are large enough to keep risk elevated.

• Liquidity: 0 safe / 1 watch / 2 danger
  Liquidity conditions are increasingly asymmetric: the system can look fine—until it doesn’t, especially around quarter-end or risk-off impulses.

• Real-Time / High-Frequency: 0 safe / 1 watch / 1 danger
  High-frequency signals aren’t screaming recession, but they’re not clean, and they’re vulnerable to shock-driven regime shifts.

---

Biggest Movers

1. ON RRP Facility ($675M): +10,617.1% (7D)
   Confirmatory (worsening tail risk). A sharp move off a near-depleted base suggests money-market plumbing is shifting; even if levels are small, the direction matters for liquidity stress sensitivity.

2. GDP Growth (QoQ annualized) (1.5%): +31.3% (7D)
   Contradictory (improving). Higher growth readings reduce near-term recession odds—though the absolute level remains “watch.”

3. NY Fed Recession Probability (0.9%): +18.7% (7D)
   Confirmatory (worsening). Even if model-based probabilities are noisy, a 7D increase aligns with other late-cycle warnings.

4. Personal Savings Rate (3.0%): +15.4% (7D)
   Contradictory (improving). A higher savings rate marginally improves household resilience—but 3.0% remains very low and still signals limited cushion.

5. Yield Curve (2s10s) (0.40): +11.1% (7D)
   Confirmatory (late-cycle risk). Continued steepening post-inversion often coincides with “late expansion” conditions rather than early expansion.

---

90-Day Indicator Trends

Using the 90-day history provided (with many series shown from mid-June through early July), the key message is not broad collapse—it’s a tug-of-war between stable coincident conditions and worsening cyclicals/fragilities.

Labor & income (stable-to-softening)

• Initial claims improved from 229K (Jun 12) → 215K (Jun 26 / Jul 1), and today remain 206K—still healthy and consistent with limited layoffs.
• Unemployment rate in the history sits around 4.3% in mid/late June, while today is 4.1% (WATCH). That’s a modest improvement versus early summer, consistent with “no recession trigger.”
• Real personal income ex transfers edged $16.5T (Jun 12) → $16.6T (Jun 26/Jul 1); today $16.6T remains “watch” mainly on momentum rather than level.

Housing (clear deterioration signal)

• Housing starts show a discrete downshift: 1465K (Jun 12) → 1177K (Jun 17 onward) in the history, and today’s 1239K (WARNING) remains below trend. Housing is still a recession-sensitive channel.
• Building permits drifted down 1423K (Jun 12) → 1410K (late June/early July); today 1433K is only modestly higher and still “watch.”

Financial conditions & credit (supportive—but “brittle”)

• Chicago Fed NFCI stayed loose around -0.51 to -0.52 in the history; today is -0.56 (SAFE)—still supportive for growth.
• High-yield OAS in the history moved 280 bps (Jun 12) → ~263–283 bps range; today ~265 bps (SAFE) remains tight—credit is not pricing recession imminently.

Markets & valuation (risk is in the divergence)

• Equity indexes in the early-summer slice rose overall:
  • S&P 500: 7394 (Jun 12) → 7499 (Jul 1); today 7719.
  • NASDAQ: 25810 (Jun 12) → 26214 (Jul 1); today 26507.
• Valuation-risk metrics stayed elevated:
  • S&P 500 / GDP hovered around 0.231–0.237 in June/early July and is 0.2376 (WARNING) today.
  • NASDAQ / GDP remained in ~0.80–0.84 (DANGER) territory in the history and is 0.8159 (DANGER) today.
• Copper-to-gold ratio sat at an extreme danger level throughout the historical slice (flat at 0.00077) and remains DANGER today—signaling persistent industrial-cycle pessimism even while equities stay buoyant.

Household buffer (slightly better, still weak)

• Personal savings rate improved from 2.6% (DANGER in June) → 3.0% (WARNING by late June/early July); today it remains 3.0% (WARNING). Direction is good; level is still a vulnerability.

Net: the 90-day trend profile supports “moderate risk”—labor/credit are not recessionary, but housing + cyclicals + valuation divergences keep the system exposed to shocks.

---

Stock Screener Signals

Today’s quant flags skew heavily toward “value dividend” profiles—ARCC, AIG, BBY, FNF, HMC, T, BCE, LTM—with a smaller cluster of “oversold growth” (CHTR, TLK) based on low RSI readings (e.g., CHTR RSI 28, TLK RSI 30).

Macro interpretation: when a screener consistently surfaces low P/E, high-yield “value dividend” names, it often reflects a market that is pricing slower growth and seeking carry, even if the index level remains high. This aligns with the macro mix we see today: tight credit spreads and low VIX (risk-on surface), paired with weak sentiment, deteriorating cyclicals, and late-cycle curve dynamics (risk under the hood).

Two important caveats for positioning signals:

• The listed dividend yields (several hundreds of percent) are almost certainly data artifacts (special dividends, stale price/dividend fields, or corporate-action distortions). Treat the factor classification (value/dividend, oversold growth) as the signal—not the raw yield numbers.
• Oversold flags in communications/telecom (CHTR, TLK) can be consistent with a market anticipating margin pressure (funding costs, competitive pricing) or defensive rotation away from growth multiples.

---

Latest Economic Developments

1) Fed policy: “knife-edge” September decision, inflation data is the hinge
In the last several days, reporting has emphasized that the Fed’s next decision may come down to very small differences in near-term inflation prints, with policymakers (and market participants) treating the upcoming inflation data as decisive for the Sep 16, 2026 meeting. (axios.com)

2) Labor data: August jobs rebound shifts focus back to inflation risk
The +162K August payrolls and 4.1% unemployment rate have been widely framed as reducing near-term recession odds while increasing the probability that the Fed stays restrictive or even tightens if inflation doesn’t cooperate. (bls.gov)

3) Growth nowcast: Atlanta Fed GDPNow remains strong (as of Sep 3)
Atlanta Fed GDPNow’s ~4.7% SAAR for 2026:Q3 (Sep 3) remains an important counterweight to recession narratives. (atlantafed.org)

4) Markets & geopolitics: oil-driven inflation anxiety returned to the tape
Recent market coverage flagged oil price strength and inflation worries as key drivers of equity pullbacks ahead of the week’s inflation reports, reinforcing that the macro regime is sensitive to energy-driven inflation re-acceleration. (apnews.com)

---

Near-Term Outlook (Next 30 Days)

The next month is a policy-and-inflation corridor: recession odds remain moderate, but the distribution is skewed—good outcomes look like “continued expansion,” while bad outcomes look like “fast tightening shock.”

Key catalysts:

• Sep 11, 2026: CPI (and Sep 11 preliminary UMich sentiment release at 10am ET)
  Michigan’s site confirms the next sentiment release date and time. (sca.isr.umich.edu)
• Sep 16, 2026: FOMC decision
  Market chatter (and Atlanta Fed market-probability tracking) suggests high odds of a hike into that meeting window, meaning risk assets and credit can reprice quickly if CPI surprises. (atlantafed.org)

What would move the risk score meaningfully higher (into the 45–55 range):

• Initial claims: sustained uptrend (e.g., multiple weeks above the low-200Ks)
• Unemployment: move that pushes the Sahm-style reading materially toward 0.50
• Credit: HY OAS widening from ~265 bps toward levels that signal stress rather than “easy conditions”
• Oil/inflation: another leg higher that forces hawkish repricing

What would move it lower (into the low-30s):

• Softer inflation prints that remove the hike tail-risk
• Stabilization in temp help and freight
• Further improvement in household buffer metrics (savings rate) without a growth slowdown

---

Long-Term Outlook (3-6 Months)

Over a 3–6 month horizon, the macro story looks like a late-cycle expansion with rising convexity:

• Base case (still most likely): growth continues, labor stays decent, credit remains functional, and recession risk stays moderate rather than high. GDPNow-style strength supports this base case in the near term. (fred.stlouisfed.org)
• Key vulnerability: the system’s “shock absorber” is thinner than it looks because:
  • Households have a low savings cushion (3.0%).
  • Cyclical “canaries” (temp help, freight, sentiment) are already soft.
  • Fiscal constraints (debt/interest expense) and banking mark-to-market issues raise the odds that a tightening in funding conditions becomes nonlinear.

Historical parallel (pattern, not prediction): late-cycle periods where equities stay near highs while cyclicals and confidence weaken can persist—until a catalyst (often inflation/energy or policy error) forces repricing. In that scenario, the recession probability can jump rapidly even if it looked benign 60–90 days prior.

---

What to Watch

Hard thresholds / triggers

• Sahm Rule: watch for movement toward 0.50 (recession trigger zone). Today: -0.07.
• Initial claims: a sustained move away from ~200K levels (trend matters more than one print).
• HY spreads (OAS): a regime shift from ~265 bps toward materially wider levels.

High-impact dates

• Sep 11, 2026: CPI + preliminary September UMich sentiment (10am ET). (sca.isr.umich.edu)
• Sep 16, 2026: FOMC decision; decision framed as “knife-edge” depending on inflation prints. (apnews.com)

Market regime tells

• Oil and inflation expectations: if energy stays bid, the Fed reaction function becomes more hawkish.
• Yield curve dynamics: continued post-inversion steepening paired with deteriorating cyclicals is the combination that elevates late-cycle risk.

---

Sources

• apnews.com
• apnews.com
• apnews.com
• apnews.com
• apnews.com
• axios.com
• axios.com
• axios.com
• bls.gov
• dol.gov
• federalreserve.gov
• federalreserve.gov
• federalreserve.gov
• kiplinger.com
• sca.isr.umich.edu
• atlantafed.org
• fred.stlouisfed.org
• atlantafed.org