Recession Risk 34/100 — July 22, 2026

Recession Risk Score: 34/100 — MODERATE (-10 vs 30 days ago)

Today’s Recession Risk Score is 34/100 (MODERATE), down 10 points from 30 days ago (44 → 34). The downgrade reflects a familiar pattern: “soft patch” signals (sentiment, temp help, freight) are still flashing, but the hard stop triggers (claims, credit spreads, insured unemployment) remain benign. Markets are behaving like the expansion is intact—equities near highs and HY spreads tight—which keeps near-term recession odds contained. The risk isn’t gone; it’s increasingly concentrated in the household and early-labor leading edge, not in broad labor or funding stress.

---

Score Trend — Last 30 Days

The score trajectory over the past 30 days shows a clear mean-reversion lower: Start 44 (Jun 22) → End 34 (Jul 22), with a 30-day min of 33 and max of 44 (average 37). In other words, the risk regime has de-escalated, but not into “all clear”—we’re still in a zone where a few breaks (claims upshift, credit widening) could pull the score back toward the low-40s quickly.

The most important feature of the path is choppiness in the last 10 readings—oscillating between the mid-30s and high-30s (34–38). That shape implies stabilization rather than a continuous improvement trend. Translation: the system is no longer pricing “imminent recession,” but it also isn’t building a durable “re-acceleration” narrative.

---

Key Drivers

1. Labor-market trigger set remains far from firing

   • Sahm Rule: 0.07 (SAFE) — well below recession-trigger territory.
   • Initial claims: ~208k (SAFE) — layoffs remain historically low, consistent with continued hiring and income support. (apnews.com)

2. Yield-curve risk has shifted from “acute warning” to “watch the re-pricing”

   • 2s10s: +37 bps (WATCH) — the curve is meaningfully re-steepened, reducing the classic near-term inversion warning signal.
   • The risk now is less “inversion implies recession” and more: whether the steepening is “good” (growth up) or “bad” (front-end cuts pricing a slowdown).

3. Leading indicators are not collapsing

   • Conference Board LEI: -0.2% m/m in June, but the six-month change is +1.1%—a notable improvement vs late-2025 dynamics. (conference-board.org)
   • This is not what “imminent recession” composites usually look like.

4. Business-cycle diffusion is still expansionary, but labor inside manufacturing is soft

   • ISM Manufacturing PMI: 53.3 (June), consistent with positive GDP growth; however Employment Index: 49.7 remains contractionary. (ismworld.org)
   • This mix fits a profit-protecting expansion (output/new orders okay; hiring cautious).

5. Financial conditions remain loose (market is not pricing stress)

   • HY OAS: ~270–273 bps (SAFE)—tight spreads are inconsistent with a near-term recession regime. (convextrade.com)
   • NFCI: -0.54 (SAFE) reinforces “easy conditions” dynamics.

6. The recession-leading “soft” cluster is still ugly

   • Temporary help (DANGER) and freight (DANGER) remain classic early warnings.
   • UMich sentiment: 44.8 (DANGER) is crisis-level pessimism—often a precursor to spending downshifts if it persists.

---

Category Breakdown

Using today’s CATEGORY BREAKDOWN counts:

• Primary Indicators: 3 safe / 4 watch / 2 danger
  Mixed. The “danger” is concentrated in leading labor/behavioral signals rather than claims-based deterioration.

• Secondary Indicators: 2 safe / 0 watch / 1 danger
  Net supportive, but not uniformly strong—still consistent with a slowing expansion.

• Housing & Construction: 0 safe / 1 watch / 1 danger
  Housing remains a dragging cyclical pocket, with permits/stops below trend—June permits were 1,367,000 SAAR (your warning-level reading). (census.gov)

• Business Activity: 2 safe / 1 watch / 0 danger
  The production side is holding together; soft patch risk hasn’t translated into a broad contraction signal.

• Consumer Credit Stress: 0 safe / 3 watch / 1 danger
  This is the underappreciated fault line: delinquencies/DSR/savings suggest the consumer is more fragile than the labor market implies.

• Market Signals: 7 safe / 2 watch / 5 danger
  A split regime: price/vol are calm (safe), but valuation and “fear ratios” (e.g., copper/gold, NASDAQ/GDP) are flashing risk.

• Liquidity: 0 safe / 1 watch / 2 danger
  Liquidity is not “crisis,” but buffer depletion matters at the margin—your ON RRP nearly depleted is consistent with less excess cash sitting at the Fed.

• Real-Time / High-Frequency: 0 safe / 1 watch / 1 danger
  High-frequency is sending “slowdown” signals (freight), but not corroborated by claims.

---

Biggest Movers

Top 5 by absolute 7-day % change (and what they mean):

1. Conference Board LEI: +673.3% (7D) — Contradictory / improving
   This magnitude is mostly a base-effect artifact (moving from a negative/near-zero baseline). Directionally, it supports “not recession-imminent,” consistent with LEI’s June profile. (conference-board.org)

2. Yield Curve (2s30s): +410.0% (7D) — Contradictory / improving near-term
   A rapid steepening typically reduces “classic inversion” recession timing risk, but it can also reflect a front-end repricing to future cuts.

3. ON RRP Facility: -99.2% (7D) — Confirmatory (liquidity erosion)
   Depletion isn’t automatically recessionary, but it reduces a cushion that helped money markets absorb shocks during prior tightening episodes.

4. NY Fed Recession Probability: -83.5% (7D) — Contradictory / improving
   A sharp drop reinforces the “markets/term structure not screaming recession” story (though probabilities can be jumpy in model outputs).

5. NASDAQ / GDP Ratio: +37.2% (7D) — Confirmatory (late-cycle valuation risk)
   This is a market fragility signal: not a recession trigger by itself, but it increases the odds that a growth scare turns into a risk-off shock.

---

90-Day Indicator Trends

Your “90-day” history window (as provided) mostly spans late April to mid-May observations for many series, so trend inference is constrained by thin sampling. Still, several meaningful direction-of-travel signals stand out:

Labor: steady on claims, but leading edges soft

• Initial claims improved from 214k (Apr 24) to ~189k (May 1), then drifted back to ~211k (mid-May)—still very low.
• The latest narrative point remains ~208k for week ending July 11 (per your summary), reinforcing “no layoffs wave.” (apnews.com)
• Sahm Rule in the history steps down 0.20 → 0.13 by mid-May—still “safe.”
• But the temp help level is in DANGER throughout (2475k → 2485k in April/May history, and your current 2,499k), keeping the “early labor downshift” signal alive.

Growth: nowcasts below trend, not collapsing

• GDPNow is shown at 1.8% across the history block; Atlanta Fed’s recent research-data snapshot has shown GDPNow updates around ~1.7% in mid-July. (atlantafed.org)
• This profile is consistent with sub-trend growth, not contraction.

Leading indicators: improved vs late-2025 regime

• Conference Board LEI: June 2026 -0.2% m/m, but six-month change +1.1%, signaling a leading-index backdrop that is not recession-consistent right now. (conference-board.org)

Credit: tight spreads are the strongest “no recession in 90 days” input

• Credit spreads in the history compress from ~320 bps to the mid-270s by early/mid-May and remain ~270–273 in mid-July. (convextrade.com)
• In past cycles, recession risk tends to rise sharply only after sustained spread widening—not tightening.

Housing: cooling remains a drag

• Your current building permits 1,367k aligns with the Census June release (July 17). (census.gov)
• Housing isn’t collapsing, but it’s not contributing strong cyclical momentum either.

Consumer: the stress is in “mood + buffers,” not yet in layoffs

• Personal savings rate falls from 4.0% → 3.6% in the history window and is 3.0% today (WARNING). That’s a thin cushion heading into any shock.
• UMich sentiment 44.8 is confirmed on FRED for May 2026 and remains a key bearish input in your dashboard. (fred.stlouisfed.org)
• Important nuance: sentiment showed a rebound in June in University of Michigan commentary (June sentiment 49.5), implying the mood trough may have been May, but today’s score still treats the household mood signal as extreme. (isr.umich.edu)

---

Stock Screener Signals

Today’s screener is dominated by “value dividend” flags—ARCC, AIG, BBY, FNF, HMC, T, BCE—plus a couple of oversold growth names (CHTR, TLK). The macro read-through is that positioning is leaning toward carry + defensiveness rather than pure cyclical beta. That aligns with the macro mix we see: growth not dead, but fragile enough that investors still prefer income and valuation discipline.

Two caveats matter:

• Several listed yields are obviously non-economic as stated (e.g., ARCC “1002%”), which suggests the screener’s yield field is likely data-scaled or malformed rather than a real forward yield. Use the style signal (value/dividend vs oversold growth) more than the yield magnitudes.
• The presence of oversold growth (e.g., CHTR with RSI 28) alongside value dividend names suggests a market that is not panic-selling, but selectively discounting idiosyncratic/levered growth exposures—consistent with a slowdown scare rather than a broad earnings collapse.

Net: the screener supports a MODERATE recession-risk regime—investors are not pricing recession, but they’re also not acting like a clean re-acceleration is underway.

---

Latest Economic Developments

Markets: U.S. equities pushed higher on Tuesday, July 21, 2026, with the S&P 500 up 0.9% per AP’s market wrap—consistent with the “financial conditions still loose” signal set. (apnews.com)

Claims and labor: The most recent confirmed claims datapoint remains 208,000 (week ending July 11, reported July 16), described as the fewest in 10 weeks—still inconsistent with recession onset. (apnews.com)

Leading indicators: The Conference Board reported the LEI fell 0.2% in June 2026, but highlighted the six-month growth at +1.1%, which is the key reason LEI is not validating the more bearish soft indicators right now. (conference-board.org)

Business surveys: ISM’s June manufacturing report shows PMI 53.3, with Prices still high but cooling (73.0 vs 82.1) and Employment still below 50 (49.7)—a signature of expansion with cautious hiring. (ismworld.org)

Fed communication: Vice Chair Jefferson’s mid-July remarks (posted/updated July 21) emphasize shock-navigation and the role of the output gap in assessing demand vs supply—consistent with a Fed that remains data dependent and not eager to introduce new policy volatility without evidence. (federalreserve.gov)
Separately, the Fed published the June 16–17, 2026 minutes on July 8, keeping the narrative anchored in “resilience but uncertainty,” rather than imminent easing. (federalreserve.gov)

---

Near-Term Outlook (Next 30 Days)

Base case: continued expansion with a growth slowdown, not recession, but with rising sensitivity to labor data.

Key catalysts over the next month:

• Weekly jobless claims (Thurs, July 23; week ending July 18) are the next immediate “tripwire” for the labor story (calendar previews highlight claims as a focus point). (kiplinger.com)
• July consumer sentiment final (July 31) could confirm whether the May collapse was the trough and whether household mood is stabilizing. (data.sca.isr.umich.edu)
• July payrolls (Aug 7, 2026) is the big one: a move in unemployment and participation will determine whether “soft patch” becomes “labor crack.”

What could move the score materially higher (toward ELEVATED):

• A sustained claims upshift (not one print)—especially if the 4-week average turns decisively up.
• HY OAS widening from ~270s into a sustained 350–400+ regime (stress repricing).
• A renewed deterioration in LEI breadth/6-month change (moving from “mixed” to “persistent decline”).

---

Long-Term Outlook (3-6 Months)

The macro picture is splitting into two narratives:

1. Hard-landing odds are restrained by labor and credit.
   The claims/Sahm/insured-unemployment cluster says the economy is not in the “recession ignition” phase. Tight HY spreads reinforce that corporate funding stress is absent—an important reason today’s score is 34, not 50+.

2. Soft-landing durability is threatened by consumer psychology and buffer depletion.
   Even if layoffs remain low, the combination of very low savings, rising consumer credit stress, and extreme sentiment can translate into a consumption downshift—especially if inflation or energy shocks reappear. The household doesn’t need mass layoffs to slow spending; it needs uncertainty + thin cash buffers.

Historically, expansions end when multiple systems fail simultaneously: labor deteriorates, credit reprices, and business activity rolls over. Today, only select leading/behavioral systems are flashing red. That’s why the correct stance is: moderate risk, tightly monitored, not “all clear.”

---

What to Watch

Labor tripwires

• Initial claims: Watch the 4-week average; a sustained climb is the earliest “hard data” confirmation of softening.
• Unemployment rate: A move from 4.2% into a persistent uptrend would raise Sahm-rule pressure quickly.

Credit tripwires

• HY OAS: sustained widening above ~350 bps would be the first regime change signal.
• Bank unrealized losses / liquidity: any signs of deposit stress or funding strain would make the “paper losses” channel macro-relevant.

Demand / household

• UMich sentiment final (July 31): confirm if sentiment is recovering (June was higher than May in UMich releases) or re-collapsing. (ng.investing.com)
• Savings rate: continued erosion from already-low levels increases the chance that a shock becomes a spending retrenchment.

Business cycle

• Conference Board LEI: watch whether the six-month change stays positive; a rollover back below zero would be a major escalation. (conference-board.org)
• ISM employment subindexes: ongoing sub-50 readings are a warning that hiring appetite is weak even with output growth.

---

Sources

• apnews.com
• macrolighthouse.com
• federalreserve.gov
• conference-board.org
• ismworld.org
• bls.gov