Recession Risk 38/100 — September 14, 2026

Recession Risk Score: 38/100 — MODERATE (+1 vs 30 days ago)

Today’s Recession Risk Score is 38/100 (MODERATE), up +1 from 37 thirty days ago (Aug 15 → Sep 14 window). The headline read remains “moderate, not imminent” because the labor-market trigger set (Sahm/claims/insured unemployment) is still firmly non-recessionary, while credit stress remains muted. The reason the score is not lower is simple: late-cycle cyclicals (temp help, freight, housing) are weak, and policy risk is rising into the Sep 16 FOMC decision as inflation stays sticky and markets price a meaningful probability of a hike.

Score Trend — Last 30 Days

The last 30 days show a gentle uptrend with sharp day-to-day oscillations: Start 37 → End 38 (Δ +1), Min 34, Max 38, Avg 36 (31 samples). The distribution matters: we’ve spent a lot of time at 34, but the score repeatedly “snaps back” to 38, suggesting the model is mean-reverting higher when market/policy inputs re-tighten.

The last 10 readings are especially telling: 34s dominate, yet Sep 6 / Sep 10 / Sep 14 print 38. That pattern is consistent with a macro backdrop where hard recession triggers are quiet, but financial conditions and cyclical leading signals keep reintroducing tail risk. In other words: no imminent contraction, but a rising sensitivity to policy and energy-driven inflation surprises heading into mid-September.

Key Drivers

1. Labor-market recession triggers remain “OFF.”

   • Initial jobless claims: 206k (week ending Sep 5; reported Sep 10) — still historically low and consistent with limited layoffs. (content.govdelivery.com)
   • Sahm Rule tracker: -0.07 (SAFE) — the cleanest real-time recession alarm is not close to triggering.

2. Growth nowcasts are expansionary—but increasingly split.

   • Atlanta Fed GDPNow (2026:Q3): 4.4% (Sep 10 update; down from 4.7% Sep 3). (atlantafed.org)
   • NY Fed Staff Nowcast (2026:Q3): 2.2% (with wide uncertainty bands). (newyorkfed.org)
     The divergence argues for “growth, but uneven”—often what you see when services hold up while goods/housing lag.

3. Policy mistake risk is climbing into Sep 16 as inflation prints hot.

   • August CPI came in hot enough to keep a September hike “live,” with energy pressures featuring prominently. (axios.com)
   • Market pricing for a hike has risen sharply in the past week per widely followed FedWatch-style coverage. (axios.com)

4. Housing is soft and vulnerable to any incremental tightening.

   • Existing-home sales weakened in August to the slowest pace in more than a year (per AP coverage), consistent with your WARNING housing starts / WATCH permits signals. (apnews.com)

5. Credit spreads are tight; broad stress remains absent.

   • HY OAS ~2.70% (tight by historical standards), consistent with no generalized credit accident yet. (alfred.stlouisfed.org)
   • Chicago Fed NFCI remains loose (~-0.56) (latest weekly reading in early September), reinforcing “risk appetite still present.” (fred.stlouisfed.org)

6. Cyclical warning cluster (temp help + freight + metals ratios) keeps recession “optionality” alive.

   • Temporary Help Services: DANGER (employment category that typically rolls before broader labor turns).
   • Freight Transportation Index: DANGER (goods economy still signaling contraction-like behavior).
   • Copper-to-gold ratio: DANGER (risk-off growth signal), though market valuation indicators complicate interpretation.

Category Breakdown

Using your CATEGORY BREAKDOWN counts:

• Primary Indicators: 3 safe / 4 watch / 2 danger
  Mixed but not recessionary overall; the key is that “danger” is concentrated in leading labor (temp help) rather than claims/Sahm.

• Secondary Indicators: 2 safe / 0 watch / 1 danger
  Secondary signals are mostly stable; the single danger print is a reminder that the “second derivative” of growth is slowing.

• Housing & Construction: 0 safe / 1 watch / 1 danger
  Housing remains the most interest-rate-sensitive pocket; recent reporting confirms demand is cooling as rates/prices bite. (apnews.com)

• Business Activity: 2 safe / 1 watch / 0 danger
  Business activity reads as slowing expansion, consistent with LEI improving modestly and profits staying healthy.

• Consumer Credit Stress: 1 safe / 2 watch / 1 danger
  Not a credit crunch, but the household buffer is thin (savings rate low), and delinquencies/DSR are worth monitoring.

• Market Signals: 7 safe / 2 watch / 5 danger
  Markets are “calm on the surface” (VIX low, indices high) but flashing valuation fragility (NASDAQ/GDP, etc.). This combination tends to mask macro vulnerability until policy/liquidity shifts.

• Liquidity: 0 safe / 1 watch / 2 danger
  Liquidity signals are where tail risk can form quickly; a near-empty RRP facility is not automatically bearish, but it can change money-market dynamics at the margin.

• Real-Time / High-Frequency: 0 safe / 1 watch / 1 danger
  High-frequency inputs are consistent with late-cycle cooling, not a collapse—yet.

Biggest Movers

Top 5 by absolute 7-day % change:

1. ON RRP Facility ($5B): -62.0% (7D)
   Interpretation: Mostly liquidity plumbing, not a direct recession trigger. Still, shrinking RRP can coincide with tighter reserves if Treasury issuance and bank balance-sheet constraints bind. Confirmatory (slightly worsening tail risk) given your liquidity category already leans danger.

2. Freight Transportation Index (-0.3): -40.0% (7D)
   Interpretation: Confirmatory (worsening risk) for the “goods recession” narrative—freight weakness aligns with soft housing/goods.

3. Sahm Rule (-0.07): -30.0% (7D)
   Interpretation: Contradictory (improving) for recession risk. Falling Sahm pressure is a strong counterweight to cyclical softness.

4. NY Fed Recession Probability (3.4%): -28.3% (7D)
   Interpretation: Contradictory (improving)—model-based recession probability is moving lower, consistent with non-recession labor signals.

5. Yield Curve (2s10s) (0.33): +16.7% (7D)
   Interpretation: Contradictory (improving near-term) since a steeper curve reduces immediate inversion signal risk—but note the historical lag: steepening can also happen when markets anticipate policy cuts later.

90-Day Indicator Trends

Your 90-day history window is partial in the excerpt (many series show June–early July snapshots), but the direction of travel across key indicators is still clear from the data you provided:

• Labor (high-frequency): claims drifting lower, not higher.
  Initial claims moved from ~229k (mid-June) to ~215k (early July) in your history, and remain ~206k in the most recent weekly report (Sep 10 release for week ending Sep 5). That’s opposite of a pre-recession pattern. (content.govdelivery.com)

• Sahm Rule improving vs early-summer levels.
  The Sahm series fell from ~0.10 (mid/late June) toward ~0.07 (early July) in your history and prints -0.07 today. That’s a meaningful improvement in the most reliable “recession trigger” family.

• Housing: step-down remains the core macro weak spot.
  Housing starts show a sharp level drop in the history (e.g., ~1465k → ~1177k), and your current reading (1239k, WARNING) suggests the sector is not re-accelerating. Recent reporting on August existing-home sales confirms demand softness into September. (apnews.com)

• Liquidity: RRP trending toward depletion.
  Your on-RRP history shows episodic spikes but a general drift lower into tiny levels, culminating in $5B now. That’s not inherently recessionary, but it increases the system’s sensitivity to rate/issuance shocks.

• Credit: spreads stay tight and stable.
  HY OAS oscillates in a narrow band (mid-260s to low-280s bps in your history) and is still around ~2.7%. That is inconsistent with impending broad default stress. (alfred.stlouisfed.org)

• Financial conditions: still loose.
  NFCI remains around -0.56 in early September—supportive of risk assets and generally inconsistent with recession onset unless labor cracks. (fred.stlouisfed.org)

• Growth nowcasts: high but susceptible to revisions.
  GDPNow is 4.4% but down from 4.7% one week earlier (Sep 3 → Sep 10). That’s still strong growth, but the direction (downward revisions) is worth monitoring into the Sep 16 update. (atlantafed.org)

Stock Screener Signals

Today’s screener is dominated by “value dividend” flags (ARCC, AIG, BBY, FNF, HMC, T, BCE, LTM) with a smaller cluster of “oversold growth” (CHTR, TLK). That mix is consistent with a market that’s not positioning for a recession outright, but is tilting toward carry and valuation discipline—the classic posture when investors expect slower growth + higher-for-longer policy risk, not a sudden demand collapse.

Two important reads beneath the surface:

• Defensive carry is winning mindshare. A screen full of “value dividend” typically signals investors want income + margin of safety while remaining invested. That aligns with your macro composite: tight spreads, loose NFCI, low VIX—risk is still being taken, but with a preference for cash-flow visibility.
• Oversold growth names show selective pressure rather than broad liquidation. CHTR (very low RSI) and TLK suggest pockets of stress, but not the “everything breaks” breadth you see when recession risk becomes imminent.

One caveat: several yields in the screener appear mechanically extreme (e.g., ARCC at 1002%). Treat these as data artifacts or corporate-action distortions rather than literal forward yields—directional takeaway (value/carry bias) still stands.

Latest Economic Developments

Inflation remains the macro headline, with energy a key accelerant. August CPI stayed hot enough to keep the Fed on a tightening path; reporting attributes much of the renewed pressure to energy dynamics and geopolitical stress. (axios.com) At the wholesale level, August PPI also ran hot year-over-year, reinforcing that pipeline inflation is not fully contained. (apnews.com)

Markets are actively repricing the probability of a September hike. Recent coverage points to a sharp increase in market-implied odds of a move at the Sep 16, 2026 FOMC decision, with the inflation prints and oil/rates interplay as the narrative driver. (axios.com) This is the single biggest “event risk” for the next 48 hours of macro trading because housing and goods are already soft.

Labor data continues to argue “slowdown, not contraction.” The official weekly claims report showed 206,000 initial claims (week ending Sep 5), consistent with layoffs remaining rare. (content.govdelivery.com) The latest employment report (Aug) showed +162k payrolls and 4.1% unemployment, signaling cooling momentum but still expansion. (bls.gov)

Housing demand is weakening at the margin. August existing-home sales fell to the slowest pace in more than a year, a reminder that housing is still the economy’s “rate-sensitivity transmission channel.” (apnews.com)

Growth trackers remain positive but fragile to incoming data. GDPNow remains strong at 4.4% for Q3, while the NY Fed nowcast is a more moderate 2.2%, implying that the growth picture depends heavily on how the next few releases land. (atlantafed.org)

Near-Term Outlook (Next 30 Days)

The next month is primarily about policy + the consumer:

• Sep 16 (Wed): FOMC decision is the defining catalyst. A hike (or a hawkish hold) that materially tightens financial conditions would likely push the score from 38 into the low-to-mid 40s quickly because housing/goods are already weak and valuation/loss sensitivity is high.
• Sep 16 (Wed): Advance retail sales for August hits the same morning as the Fed decision day and will help determine whether consumption is re-accelerating or rolling over. (fred.stlouisfed.org)
• Weekly claims remain the cleanest near-term tripwire: if claims break above ~240–260k and stay there, your “moderate” call shifts toward elevated quickly.
• Inflation follow-through: PPI/CPI hot prints raise the odds of additional hikes later in 2026, even if September is a hold—keeping the “policy mistake” channel open.

Base case for the next 30 days: moderate risk persists, with the score oscillating between mid-30s and high-30s unless a clear claims uptrend or abrupt credit widening appears.

Long-Term Outlook (3-6 Months)

Over the next 3–6 months, the economy still looks like a late-cycle expansion with uneven sectoral performance:

• Labor is the anchor. As long as claims stay low and the Sahm Rule remains safely below trigger, a near-term recession is unlikely. (content.govdelivery.com)
• Housing/goods are the drag. The longer housing stays soft (starts/permits and weak existing-home sales), the more the slowdown can spread into employment categories that lag. (apnews.com)
• Credit is the swing factor. Tight HY spreads (~2.7%) imply investors see limited default risk; if that changes (toward ~350 bps+), recession odds rise quickly because refinancing costs and risk appetite transmit fast. (alfred.stlouisfed.org)
• Valuations and banking system sensitivity matter more in a “higher-for-longer” world. Your dashboard flags elevated valuation ratios and large unrealized losses; those aren’t recession triggers by themselves, but they increase the odds that a shock (rates, liquidity, geopolitics) becomes nonlinear.

Net: the 90-day trajectory you provided points to stable expansion in labor + tight credit, but structural fragility in cyclicals and policy sensitivity. That keeps recession risk moderate rather than low.

What to Watch

Event risk / calendar

• Sep 16, 2026: FOMC rate decision + press conference (policy path clarity).
• Sep 16, 2026: Advance retail sales (August) for evidence of consumer rollover. (fred.stlouisfed.org)

Thresholds that would change the score fast

• Initial claims: sustained move >240k–260k (and especially accelerating).
• Credit spreads: HY OAS widening from ~270 bps → 350+ bps.
• Yield curve: re-inversion and/or abrupt bear steepening tied to inflation fears.
• Housing: further deterioration in starts/permits plus additional evidence of sales contraction.

Confirmation signals

• Temp help: stabilization would reduce recession optionality; continued declines would be an early warning that the labor slowdown is propagating.
• Freight: improvement would argue the goods recession is ending; continued weakness supports a “slowdown spreads” narrative.

Sources

• apnews.com
• apnews.com
• apnews.com
• apnews.com
• apnews.com
• axios.com
• axios.com
• axios.com
• kiplinger.com
• kiplinger.com
• atlantafed.org
• atlantafed.org
• newyorkfed.org
• content.govdelivery.com
• bls.gov
• dol.gov
• conference-board.org
• fred.stlouisfed.org
• alfred.stlouisfed.org
• fred.stlouisfed.org
• bls.gov
• dol.gov
• dol.gov
• fred.stlouisfed.org