Recession Risk 38/100 — July 21, 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 (start 34 on June 21, 2026 to 38 on July 21, 2026). The core message remains “slowdown, not break”: top-tier labor-confirmation triggers (claims/Sahm) are still calm, even as forward cyclicals (temp help, freight, sentiment) point to late-cycle fragility. Financial conditions and credit pricing continue to argue against an imminent recession, but the balance of risks is drifting higher as growth decelerates and household buffers thin.

Score Trend — Last 30 Days

Over the last 30 days, the score has been choppy but upward-biased: Start 34 → End 38 (+4), with a min of 33 and a max of 44 (average 37, 31 samples). The pattern looks like a late-cycle “two-steps-forward, one-step-back” grind higher, rather than a clean breakout into high-risk territory.

The last 10 readings reinforce that view: we’ve repeatedly slipped back to 34 (July 15–16 and July 19) but quickly reverted to the high-30s (38 on July 12, July 18, July 20, and July 21). That “mean reversion toward the high-30s” implies the economy is not collapsing, yet the market/labor cushion is being offset by weak forward indicators that keep pulling the score back toward MODERATE.

Key Drivers

1. Labor market confirmation is still absent (risk-reducing)

   • Initial jobless claims: 208K (week ending July 11, 2026)—low and consistent with continued expansion bias. (apnews.com)
   • Sahm Rule: 0.07 (SAFE)—well below typical trigger dynamics and consistent with “slowdown without labor break.”

2. Fed policy: restrictive-but-stable (risk-neutral to slightly risk-reducing)

   • The Fed held the target range at 3.50%–3.75% on June 17, 2026, reducing near-term “policy surprise” risk. (federalreserve.gov)
   • Stability matters: in this regime, the recession risk impulse comes less from incremental hikes and more from lagged tightening + growth deceleration.

3. Growth deceleration is real (risk-increasing)

   • June payrolls: +57K—a clear downshift in hiring momentum. (bls.gov)
   • GDPNow has recently implied sub-trend growth; as of July 16, the Atlanta Fed model estimated Q2 real GDP growth at 1.7%, up from 1.3% on July 8. (atlantafed.org)
   • Net: growth isn’t screaming recession, but it’s too soft to comfortably absorb shocks.

4. Forward cyclicals are flashing amber-to-red (risk-increasing)

   • Temporary Help Services: 2,499.2K (DANGER)—temp help is a classic “first fired” labor category and is already weak. (bls.gov)
   • Freight Transportation Index: 0.3 (DANGER)—goods-side weakening continues (a common early-cycle-to-late-cycle crack).

5. Credit and financial conditions are not confirming recession (risk-reducing)

   • HY OAS: 273 bps (SAFE)—still tight, inconsistent with imminent broad default stress. Recent prints in mid-July were in the ~2.7% area. (fredaccount.stlouisfed.org)
   • Chicago Fed NFCI: -0.54 (SAFE)—financial conditions appear loose, supporting risk assets and cushioning the economy.

6. Household stress signals are building at the margins (risk-increasing)

   • Credit card delinquency: 2.9% (WATCH) and personal savings rate: 3.0% (WARNING) point to thinner buffers even if layoffs remain low.
   • This is the “slow-burn” channel: consumption softens first, then labor follows if it persists.

Category Breakdown

• Primary Indicators: 3 safe / 4 watch / 2 danger
  Mixed: labor-confirmation components are mostly safe, but the watch/danger balance indicates macro momentum is deteriorating even if recession confirmation is not present.

• Secondary Indicators: 2 safe / 0 watch / 1 danger
  Still supportive overall, but the presence of a danger signal here reinforces that weakness isn’t confined to one niche series.

• Housing & Construction: 0 safe / 1 watch / 1 danger
  Housing is no longer a tailwind. Permits (warning) and starts (watch) suggest residential activity is cooling, which often shows up before broader employment rolls.

• Business Activity: 2 safe / 1 watch / 0 danger
  Business-side “current activity” looks okay, but the watch count says momentum is what’s at risk, not the current level.

• Consumer Credit Stress: 0 safe / 3 watch / 1 danger
  This is a meaningful yellow flag: stress is not acute, but it’s broadening (delinquencies, debt service, savings).

• Market Signals: 7 safe / 2 watch / 5 danger
  A split personality: index levels/volatility are calm, but several valuation and macro-ratio indicators are extreme. This combination can mask late-cycle fragility until it doesn’t.

• Liquidity: 0 safe / 1 watch / 2 danger
  Liquidity plumbing is less forgiving (e.g., very low RRP), which can amplify volatility if a catalyst hits.

• Real-Time / High-Frequency: 0 safe / 1 watch / 1 danger
  High-frequency data are not screaming recession, but they are no longer cleanly expansionary.

Biggest Movers

1. Conference Board LEI: +673.3% (7D) — contradictory / improving
   A large % move likely reflects a base effect or signal flip (your history shows LEI toggling between -0.3 and +1.7). Directionally, it reduces risk.

2. Yield Curve (2s30s): +410.0% (7D) — contradictory / improving (near-term)
   Steepening generally reduces immediate recession odds, but historically it can also steepen because growth expectations fall (bear steepening vs bull steepening matters). Treat as near-term risk relief, not “all clear.”

3. NY Fed Recession Probability: -83.5% (7D) — contradictory / improving
   A sharp drop in that model output is supportive of the “not imminent” view.

4. SLOOS Lending Standards: -19.0% (7D) — contradictory / improving
   Less tightening is a modest tailwind. The magnitude isn’t huge, but direction helps.

5. ON RRP Facility: -16.3% (7D) — confirmatory / worsening (liquidity)
   Less cash parked at RRP can reflect liquidity redistribution; in your framework it’s flagged WARNING/DANGER. This raises plumbing risk if combined with funding stress elsewhere.

90-Day Indicator Trends

Your 90-day history window (as provided) is concentrated in late April through mid-May for many series, so the “90-day” read here is best interpreted as direction-of-travel from the earliest available prints to current (today’s snapshot) plus the intermediate checkpoints embedded in the history.

Labor: still okay, but cooling at the edges

• Initial claims (available history): roughly ~207K–214K in late April, dipping to 189K (May 1), then drifting back toward ~211K by mid-May, and now 208K for the July 11 week. Net: stable low plateau, not a breakout.
• Sahm Rule: 0.20 in late April → 0.13 by mid-May → 0.07 today. That’s the opposite of recession confirmation; it’s a clear non-trigger trend.

Growth/activity: “slow, not recession”

• Industrial production: 101.8 (Apr) → 102.5 (mid-May) → 102.6 today: modest improvement in level, consistent with a “still expanding” production baseline.
• GDPNow: pinned around 1.8% in your history snapshot; as of July 16, the Atlanta Fed commentary showed 1.7% (Q2) after 1.3% on July 8. (atlantafed.org)
  Interpretation: sub-trend, not contraction, but not enough momentum to dissipate risk.

Housing: cooling

• Building permits drifted down in the history: 1386K (Apr 22) → 1372K (Apr 30) → 1363K (mid-May), and your current read 1367K remains in that softer band. Housing is not collapsing, but it’s not accelerating either.

Consumer: buffer erosion

• Personal savings rate fell from 4.0% (late Apr) → 3.6% (May) and is 3.0% today. That’s a material deterioration in cushion over a short horizon.
• Consumer sentiment in the history fell from 56.6 (Apr 24) → 53.3 (late Apr–May), while your current reading is 44.8 (DANGER) (crisis-level pessimism). Even allowing for sampling timing, the direction is clearly negative.
  Note: Michigan sentiment in public releases showed May at 44.8 and June at 49.5. (data.sca.isr.umich.edu)
  In other words, “very low sentiment” is real, and even rebounds are occurring from a deeply depressed base.

Credit/financial conditions: still supportive

• HY OAS in your history oscillates but remains in a tight regime (high-200s to low-300s bps). Mid-July readings around 2.73% support your “credit not pricing imminent recession” thesis. (fredaccount.stlouisfed.org)
• NFCI stayed around -0.52 in your history: loose conditions persist.

Markets/valuation: risk is structural, not cyclical (yet)

• Equities in your 90-day history rose meaningfully (e.g., S&P 500 ~7064 (Apr 22) → ~7409 (mid-May) → 7458 today). That’s consistent with “risk-on,” but your valuation/risk ratios (NASDAQ/GDP danger; S&P500/GDP warning) imply low margin for error.

Stock Screener Signals

Today’s flagged names cluster into two regimes: (1) “value dividend” defensives/financials, and (2) a smaller pocket of “oversold growth.” In recession-risk terms, that mix usually signals barbell positioning—investors want carry/quality balance sheet exposure, but they’re also selectively hunting drawdowns.

The value-dividend list (e.g., ARCC, AIG, BBY, FNF, HMC, T, BCE) leans toward cash-flow and yield. That can be consistent with a MODERATE-risk macro: not a full recession hedge (which would usually scream long-duration Treasuries and deep defensives), but a preference for income + valuation support as growth slows. One caution: the stated yields in your screener output look mechanically exaggerated (likely a data mapping issue), so treat the “yield factor” directionally—these are perceived income/value exposures—rather than literally.

The oversold growth signals (CHTR, TLK) suggest pockets where the market is already pricing stress (e.g., idiosyncratic leverage/regulatory/competitive issues). In a benign macro, oversold growth can mean-revert. In a deteriorating macro, these are often the first places where tight financial conditions (if they arrive) would bite. Net: the screener fits today’s macro read—not panic, but selective caution and valuation awareness.

Latest Economic Developments

Labor market: claims remain low. For the week ending July 11, 2026, initial jobless claims fell to 208,000, a level associated with a still-healthy labor market. (apnews.com) This lines up with your SAFE classification for claims and explains why the score isn’t higher despite weak forward cyclicals.

Payrolls: hiring has cooled sharply. The June 2026 Employment Situation showed +57,000 payroll gains—well below typical expansion norms—and reinforces that the economy is running at reduced momentum. (bls.gov) When hiring slows to this degree, the economy becomes more sensitive to negative surprises (energy, geopolitics, credit, policy).

Fed: on hold, reducing shock risk. The Fed’s June 17, 2026 statement maintained the funds rate target range at 3.50%–3.75%. (federalreserve.gov) Holding steady lowers near-term “rate surprise” recession risk, but the tradeoff is that policy remains restrictive enough that the lagged effects can continue to accumulate.

Growth nowcasting: sub-trend but not collapsing. The Atlanta Fed’s GDPNow commentary (as of July 16) put Q2 growth at 1.7%, up from 1.3% on July 8. (atlantafed.org) That’s consistent with a “slowdown narrative” rather than a sudden stop.

Markets: calm surface, concentrated volatility. Recent sessions have been relatively quiet, with the AP describing a drift day on July 20, 2026 as AI-related stocks trimmed some losses and broader indexes moved modestly. (apnews.com) This matches the SAFE VIX reading and supports the view that financial conditions are not currently enforcing discipline on the real economy.

Near-Term Outlook (Next 30 Days)

Base case for the next month: risk stays MODERATE (mid-to-high 30s) unless labor weakens. The key is whether the economy’s “soft hiring” turns into “job loss” signals.

Catalysts that could push the score higher (toward elevated):

• Initial claims: a sustained move from ~200K toward the mid-200Ks (and staying there) would be the fastest high-frequency confirmation.
• Unemployment / Sahm Rule: if the 3-month unemployment average accelerates, the Sahm signal can flip quickly.
• Credit spreads: HY OAS widening from ~high-200s into the mid-300s+ would indicate propagation from “soft data” to financing stress.

Catalysts that could pull the score lower (toward low-30s):

• Stabilization in temp help and freight (leading cyclicals),
• Firming in sentiment,
• Continued benign credit + steady claims.

Long-Term Outlook (3-6 Months)

Three-to-six months out, the macro picture is best described as late-cycle fragility with asymmetric downside. The strongest recession-negating evidence remains labor (claims/Sahm) and credit (tight spreads, loose NFCI). But the 90-day direction of travel in several leading/fragile areas—temp help weakness, freight deterioration, low savings, depressed sentiment—suggests the economy is losing resilience.

Historically, recessions tend not to begin because sentiment is bad; they begin when sentiment + leading cyclicals are followed by labor deterioration and credit tightening. Today, we’re still early in that chain: the first links are visible, the latter links are not. That implies the most likely path is a prolonged slowdown—with a meaningful probability of recession if (and only if) labor cracks or credit reprices.

What to Watch

Labor (highest priority)

• Initial jobless claims: watch for a regime shift above ~230K–250K and persistence.
• Sahm Rule: still SAFE; any rapid climb matters.

Growth

• GDPNow: if nowcasts drift toward ~0%, the “slowdown” becomes “stall speed.”

Credit & liquidity

• HY OAS: widening above ~350 bps would be a meaningful stress confirmation.
• Bank unrealized losses + liquidity plumbing (RRP depletion, funding stress proxies): these are tail risks that can become catalysts.

Housing

• Permits and starts: continued drift lower would reinforce late-cycle slowdown.

Markets

• Valuation extremes (NASDAQ/GDP danger): in a slowing economy, richly valued segments are vulnerable to earnings disappointments and multiple compression.

Sources

No data available for this window.