Recession Risk 34/100 — September 20, 2026

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

Today’s Recession Risk Score is 34/100 (MODERATE), and it has held steady versus 30 days ago. The macro picture remains a “two-speed” mix: real-time labor is still expansion-consistent, while goods-sensitive leading signals (temp help, freight, housing) continue to sag. The biggest incremental headwind is policy—the Fed hiked 25 bps on September 16, 2026, tightening conditions at the margin and increasing the odds that “cooling” becomes “cracking” if demand rolls over. Net: not an imminent recession setup, but the economy is late-cycle enough that small shocks can propagate.

Score Trend — Last 30 Days

Over the last 30 days (window 2026-08-21 → 2026-09-20), the score started at 34 and ended at 34 (Δ: +0), with a min of 34, max of 38, and average of 36. That’s a classic range-bound “moderate risk” regime: the system repeatedly flirts with a higher-risk zone but fails to sustain it.

The shape matters: the pattern is spiky rather than trending—brief jumps to 37–38 (notably Sep 12, Sep 14, Sep 18) followed by quick mean reversion back to 34. That typically implies conflicting inputs: some indicators are flashing late-cycle stress, but credit and high-frequency labor data keep pulling risk back down. The September 16 rate hike likely raises the probability of future spikes; the key question is whether those spikes start to “stick” (i.e., the floor rises from 34 toward the high 30s).

Key Drivers

1. Labor recession triggers remain safely off

   • Sahm Rule: -0.07 (SAFE) — still well below the 0.50 trigger and not signaling a labor-market recession impulse.
   • Initial jobless claims: 196k (SAFE) — extremely low, consistent with limited layoff pressure.
   • Interpretation: the labor market is cooling around the edges, but it is not breaking.

2. Policy tightened—Fed hike raises “policy shock” probability

   • The Fed raised the target range by 25 bps to 3.75%–4.00% on September 16, 2026. (federalreserve.gov)
   • The policy statement/SEP messaging has been interpreted as leaving the door open to additional tightening. (axios.com)
   • Translation: even if the economy is fine today, the distribution of outcomes gets worse when the Fed is still hiking late-cycle.

3. Growth nowcast is strong, but your tracker’s growth signal is softer

   • Atlanta Fed GDPNow: 5.1% SAAR for 2026Q3 (Sep 17). (atlantafed.org)
   • Your internal GDPNow tracker reading: 1.8% (WATCH) and GDP growth: 1.5% QoQ annualized (WATCH) suggest a more moderate baseline.
   • Key point: the direction of travel is what matters next—if GDPNow holds up while hiring softens, risk stays moderate; if GDPNow rolls over after the Fed hike, risk rises quickly.

4. Goods-economy leading weakness persists (yellow-to-orange flags)

   • Temporary Help Services: 2520K (DANGER) — historically a high-signal leading indicator when it declines persistently.
   • Freight Transportation Index: -0.3 (DANGER) — consistent with softening goods demand and weaker industrial momentum beneath headline strength.

5. Housing remains a clear weak spot

   • Building permits: 1,394K (WARNING) and housing starts: 1,275K (WARNING) in your dashboard.
   • Census’ latest release (Aug data, released Sep 17, 2026) confirms permits around 1.394M SAAR. (census.gov)
   • Mechanism: housing weakness rarely stays isolated if rates remain restrictive; it bleeds into durable goods, construction employment, and local services.

6. Financial conditions and credit are not recessionary

   • HY OAS ~270 bps (SAFE) — tight spreads do not corroborate imminent contraction.
   • Chicago Fed NFCI: -0.56 (SAFE) and VIX: 15.4 (SAFE) — liquidity/volatility regime is complacent-to-benign.

Category Breakdown

Using your CATEGORY BREAKDOWN counts:

• Primary Indicators: 3 safe / 4 watch / 2 danger
  Mixed: labor triggers are mostly safe, but the late-cycle labor cooling and select leading components keep primary signals from turning decisively green.

• Secondary Indicators: 2 safe / 0 watch / 1 danger
  A small set, but it’s notable that secondary is not broadly deteriorating—no reinforcement yet.

• Housing & Construction: 0 safe / 0 watch / 2 danger
  Housing is the cleanest persistent weakness in the stack; it remains the most plausible channel for tightening to transmit into the real economy.

• Business Activity: 2 safe / 1 watch / 0 danger
  Business activity is soft but not recessionary, consistent with “slowdown” rather than “downturn.”

• Consumer Credit Stress: 1 safe / 2 watch / 1 danger
  Consumer balance sheet resilience is thinning: delinquencies and debt service are creeping up while savings are low.

• Market Signals: 6 safe / 3 watch / 5 danger
  Markets are sending a split message: index levels/vol are calm, but valuation and cross-asset “fear ratios” flag vulnerability if growth disappoints.

• Liquidity: 0 safe / 1 watch / 2 danger
  Liquidity is a watch-zone: as the system relies more on private credit and market-based finance, liquidity regime changes can create fast nonlinear risk.

• Real-Time / High-Frequency: 0 safe / 1 watch / 1 danger
  High-frequency data is not flashing recession, but it’s sensitive to policy and can turn quickly if layoffs pick up.

Biggest Movers

Top 5 by absolute 7-day % move (and what it means):

1. ON RRP Facility ($5B): -45.6% (7D)

   • Confirmatory for tighter liquidity plumbing risk: less RRP usage can reflect shifting money-market dynamics; by itself not recessionary, but it can amplify volatility if reserves become scarce.

2. NY Fed Recession Probability (0.9%): -21.3% (7D)

   • Contradictory (improving): model-implied recession odds eased over the week, consistent with benign spreads/vol.

3. Yield Curve (2s30s) (0.81): +10.4% (7D)

   • Generally contradictory (improving): a steeper curve (after inversion) often aligns with improving forward growth expectations, though steepening can also occur from rising term premium.

4. VIX (15.4): -3.7% (7D)

   • Contradictory (improving): lower implied volatility signals markets are not pricing near-term macro stress.

5. Unemployment Rate (4.1%): -2.3% (7D)

   • Contradictory (improving) on the margin: lower unemployment reduces recession probability, but the bigger story is still quits/temp help vs claims.

90-Day Indicator Trends

Your 90-day history (provided) shows a key theme: the recession-sensitive leading indicators are weak, but the “hard-stop” recession triggers haven’t engaged. Because the history block you provided contains daily snapshots for many series that are effectively flat through late June/early July, the most actionable 90-day read is directional and cross-sectional rather than month-to-month volatility.

Notable changes comparing “then” (late June/early July) vs “now” (Sep 20 readings):

• Industrial Production: 102.6 → 103.1 (SAFE)
  Directionally higher (~+0.5 index points from late June), consistent with continued expansion in output.

• Sahm Rule: 0.10 → -0.07 (SAFE)
  Improvement of 0.17 points from late June. This is a big reason the headline risk score refuses to trend higher: the labor-market recession trigger is moving away from danger, not toward it.

• Initial claims: ~226k → 196k (SAFE)
  Down about 30k from late June levels in your history block. That is not a recession setup.

• Real personal income ex transfers: $16.5T → $16.6T (WATCH)
  Slight improvement, but still flagged watch—income growth may be positive, yet not accelerating enough to provide a cushion against restrictive policy.

• Temporary help services: ~2490K → 2520K (DANGER)
  Still danger. Even if the level ticked up versus late June, the classification and narrative are unchanged: temp help is behaving like a late-cycle canary.

• Housing: starts/permits remain below trend (warning/danger by your category counts).
  Census confirms permits around 1.394M SAAR in the latest release; housing remains one of the most rate-sensitive weak links. (census.gov)

• Credit spreads: ~263–283 bps range → ~270 bps (SAFE)
  Stable/tight. Credit is not validating the freight/temp-help stress—yet.

• Risk sentiment/vol: VIX ~18s → 15.4 (SAFE)
  Falling vol is inconsistent with an imminent recession but can also be late-cycle complacency if policy risk is rising.

Bottom line on 90 days: macro stress is concentrated (housing, temp help, freight, household cushion) rather than broad-based. That concentration is consistent with a moderate risk score, not elevated—until/unless labor demand (claims/continuing claims, unemployment acceleration) turns.

Stock Screener Signals

Today’s quant flags cluster in two buckets: (1) “value dividend” names with low P/E, and (2) a smaller set of “oversold growth” names with low RSI.

First, the value/dividend cluster—ARCC, AIG, BBY, FNF, HMC, T, LTM, BCE—looks like the market rewarding carry and cash flow while still being selective about cyclicality. In a “moderate risk” macro regime, this pattern often reflects positioning for:

• Slower growth but not recession (investors want yield and valuation support),
• Higher-for-longer policy risk (dividend and low multiple as ballast),
• Credit selection (ARCC-style exposure tends to benefit when default risk is contained, but can suffer if spreads gap wider).

Second, the “oversold growth” flags—CHTR (RSI 28) and TLK (RSI 30)—suggest tactical mean-reversion demand in pockets that have been punished (idiosyncratic or rate-sensitive). That aligns with the macro signal mix: tight HY spreads + low VIX is a friendly backdrop for opportunistic re-risking, but the presence of heavy “value dividend” skew indicates investors are not fully embracing a high-beta expansion narrative.

One red flag in your screener output: several listed “yields” are clearly non-economic (e.g., 1002%, 654%). Treat those as data anomalies rather than market signals; the real signal is the clustering around low multiples and defensive carry.

Latest Economic Developments

• Federal Reserve (policy): The Fed raised the federal funds target range to 3.75%–4.00% on September 16, 2026 and reaffirmed a restrictive stance to push inflation down. (federalreserve.gov) Reporting around the meeting emphasized a bias toward further tightening later in 2026. (axios.com)
  Market implication: rates-sensitive sectors (housing, long-duration equities) become more fragile, and the risk of an “accidental tightening” rises.

• Growth nowcast: Atlanta Fed GDPNow estimated 5.1% SAAR real GDP growth for 2026Q3 as of September 17. (atlantafed.org)
  Macro implication: The economy is not behaving like it’s sliding into contraction right now, which is consistent with the low claims reading and tight spreads.

• Housing (latest release): The Census Bureau release dated September 17, 2026 reported building permits at a 1.394M SAAR (August). (census.gov)
  Macro implication: housing is still running below prior-cycle norms, and any additional rate pressure risks further drag into year-end.

• Markets around the hike: Coverage noted equities softened after the decision as the Fed emphasized inflation risks and left the door open to more tightening. (apnews.com)
  Macro implication: financial conditions can tighten via expectations even when spreads/vol stay contained.

Near-Term Outlook (Next 30 Days)

Base case: risk score stays in the low-to-mid 30s unless labor data deteriorates. The Fed hike on September 16, 2026 means the next month is about transmission—whether higher short rates and firmer long rates begin to show up in:

• Claims/continuing claims (the fastest labor signal),
• Hiring/quit behavior (already weak in quits at 1.9%),
• Housing activity (already weak in permits/starts),
• Consumer stress (low savings rate at 3.0%, delinquencies elevated).

Key calendar catalysts (next ~30 days):

• Additional weekly jobless claims prints (watch for a sustained break above ~230k and then 250k).
• Inflation releases (CPI/PCE) that could harden or soften the Fed’s “one more hike” posture.
• Housing prints that confirm whether August was stabilization or continued slide.
• Earnings / guidance (especially cyclicals, consumer discretionary, and regional banks) for real-time read-through on demand and credit quality.

Long-Term Outlook (3-6 Months)

Three forces will determine whether “moderate” becomes “elevated” by year-end:

1. Policy path vs labor-market resilience
   The Fed has restarted hiking late-cycle. (federalreserve.gov) If the committee follows through with another hike later in 2026, the probability rises that labor cooling (quits/temp help) turns into layoffs (claims) with a lag. The Sahm Rule is currently safe, but it can move quickly once unemployment accelerates.

2. Housing as the transmission channel
   Housing is already weak; a prolonged restrictive stance tends to lengthen and deepen housing slowdowns. Historically, housing weakness doesn’t guarantee recession, but it raises recession elasticity—small shocks elsewhere become more damaging.

3. Credit and liquidity as “nonlinear amplifiers”
   Today, HY OAS at ~270 bps is calm. But if profits slow, refinancing costs rise, or a liquidity event hits (especially with banking unrealized losses flagged in your dashboard), spreads can widen quickly—turning a contained slowdown into a broader contraction.

My 3–6 month read from your signals: the economy looks more like a late-cycle expansion with pockets of fragility than a pre-recession cascade. The setup changes if we see (a) a persistent uptrend in initial/continuing claims, (b) a clear re-widening of HY spreads, and (c) housing rolling over further into fall/winter.

What to Watch

Labor (highest weight)

• Weekly initial claims: watch for a sustained uptrend (e.g., several prints >230k, then >250k).
• Unemployment rate acceleration: the threshold is not 4.1% itself; it’s the speed of increase and whether it pushes the Sahm Rule toward 0.50.

Credit / risk

• HY OAS: a move from ~270 bps to >400 bps would be a meaningful regime shift.
• VIX: persistent >20 would signal tightening risk appetite.

Housing

• Permits and starts: continued weakness into October/November would reinforce that policy transmission is working—raising recession odds into early 2027.

Leading/goods

• Temporary help: stabilization would be an important “false alarm” tell; renewed decline would be confirmatory.
• Freight: watch for continued negative readings—goods contraction can bleed into broader hiring.

Sources

No data available for this window.