Recession Risk 38/100 — August 20, 2026

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

Today’s Recession Risk Score is 38/100, keeping the outlook in the MODERATE band and unchanged versus 30 days ago. The data continue to describe a two-speed economy: the labor market and credit conditions remain broadly supportive, while household buffers and several cyclical “early warning” indicators (temp help, freight, and copper/gold) look fragile. The net result is a slowdown-with-tail-risk profile rather than a clean recession setup. The next 30 days will be decided by whether labor-market cooling stays contained and whether the Fed’s “higher-for-longer / conditional hike” bias collides with weakening demand.

Score Trend — Last 30 Days

The score has mean-reverted and stabilized over the last month: Start 38 → End 38 (Δ 0), with a range of 34–44 and an average near 37. This is not a grind higher (which would imply broad-based deterioration); it’s a choppy plateau where “risk-on” market inputs repeatedly offset consumer and cyclical stress.

The last 10 readings show a sawtooth pattern between 34 and 38, implying the system is sensitive to incremental news flow rather than trending in one direction. In practical terms: recession odds are not accelerating, but the economy is also not rebuilding margin-of-safety—so the score can still jump quickly if a single pillar (claims, spreads, or housing) breaks trend.

Key Drivers

1. Labor market still “safe” on the fastest signals

   • Initial claims: 209k (week ended Aug 8, 2026) remains firmly inside a historically low band. A recent Labor Department update (reported Aug 13) emphasized that claims have stayed roughly ~200k–230k for about a year, consistent with low layoffs. (apnews.com)
   • Sahm Rule: -0.03 (SAFE) confirms no unemployment-driven recession trigger is forming.

2. Yield-curve normalization reduces near-term recession probability (but watch the “cuts later” interpretation)

   • 2s10s: +46 bps (WATCH) and 2s30s: +1.06 (WATCH) are supportive near-term because the classic inversion signal has reversed.
   • However, steepening after inversion can also reflect markets pricing future easing into slower growth, so it’s a “supportive but not all-clear” signal.

3. Credit is calm: spreads tight and conditions loose

   • HY OAS ~267–271 bps in mid-August remains tight (low stress), historically inconsistent with imminent recession. (dollarliquidity.com)
   • Chicago Fed NFCI about -0.55 is loose financial conditions—another reason recession risk is not elevated right now. (fred.stlouisfed.org)

4. Consumer demand is wobbling as buffers look thin

   • Your tracker flags UMich sentiment in DANGER and personal savings rate at 2.7% (DANGER)—a classic “shock-absorption” problem.
   • Recent reporting shows July retail sales fell 0.6% (largest decline since May 2025 in that report) with core retail sales also down—evidence that the consumer can “pause” abruptly when confidence is this weak. (apnews.com)

5. Fed stance is a growth risk at the margin

   • The Fed held the policy rate at 3.50%–3.75% on July 29, 2026. (federalreserve.gov)
   • Minutes released Aug 19, 2026 showed many officials saw a scenario where higher rates might be needed if inflation doesn’t cool, reinforcing “hawkish conditionality” rather than a clean easing cycle. (apnews.com)
   • That matters because conditional hawkishness into slowing consumption is exactly how soft landings turn into hard ones.

Category Breakdown

Using your CATEGORY BREAKDOWN counts:

• Primary Indicators (3 safe / 5 watch / 1 danger): Mixed but not breaking—labor is still the anchor, while growth momentum indicators remain “monitor closely.”
• Secondary Indicators (2 safe / 0 watch / 1 danger): The “secondary” set is mostly okay, but the single danger reading is a reminder that weak sub-signals can lead the headline economy.
• Housing & Construction (0 safe / 1 watch / 1 danger): Housing is not a tailwind—permits are only watch-level, and starts are danger-level in your system, consistent with a slowing impulse from residential construction.
• Business Activity (2 safe / 1 watch / 0 danger): The business activity complex is still supportive, aligning with your “slowdown, not recession” base case.
• Consumer Credit Stress (1 safe / 2 watch / 1 danger): Early stress is present (delinquencies and debt-service pressure), but it’s not yet broad enough to force a recession call.
• Market Signals (7 safe / 2 watch / 5 danger): Markets are simultaneously easy financial conditions and valuation/defensive-warning extremes—this is the “risk-on prices, risk-off internals” split.
• Liquidity (0 safe / 1 watch / 2 danger): Liquidity is the most fragile family (RRP depletion, etc.), raising vulnerability to a tightening shock.
• Real-Time / High-Frequency (0 safe / 1 watch / 1 danger): The fast data are the tripwire—if claims, temp help, or freight worsen further, the score can reprice quickly.

Biggest Movers

From your BIGGEST MOVERS (7-day % change) list:

1. Bank Unrealized Losses: +931.1% (7D) — Confirmatory (worsening risk)
   A jump of this magnitude is a balance-sheet fragility warning: even if conditions are calm today, latent losses increase the system’s sensitivity to funding stress.

2. ON RRP Facility: -95.1% (7D) — Confirmatory (worsening risk)
   Rapid depletion reduces the “cash buffer” in the plumbing. In a benign scenario it’s just normalization; in a stress scenario it’s one less shock absorber.

3. Sahm Rule: -23.1% (7D) — Contradictory (improving)
   The Sahm reading moving down is consistent with your broader conclusion: no labor-market-triggered recession signal yet.

4. VIX: +20.2% (7D) — Confirmatory (worsening risk, mild)
   Volatility is still low in absolute terms, but the direction matters: it signals growing sensitivity to macro catalysts (Fed, inflation, consumer).

5. NY Fed Recession Probability: -19.0% (7D) — Contradictory (improving)
   Falling model-implied recession probability is consistent with the positive yield-curve slope and relatively benign credit.

90-Day Indicator Trends

Your 90-day history provided here is partial (many series show May–June snapshots), but it still highlights important direction of travel:

Labor & labor-leading

• Initial claims: 209k (May 22) → 215k (May 29) → 225k (Jun 5–11). That’s a +16k rise from late May to early June—still “safe,” but the slope is the key. A continued drift toward 240k–260k would likely lift the score meaningfully.
• JOLTS quits rate: 2.0% → 1.9% in early June (watch → warning in your labeling). Quits softening typically signals reduced worker bargaining power and slower wage churn—good for inflation, but often late-cycle.
• Sahm Rule: 0.13 → 0.10 by early June in your history (and -0.03 today). This is consistent with “labor cooling but not cracking.”

Household income, buffers, and consumer psychology

• Real personal income ex-transfers: $16.7T → $16.5T by late May/early June in your data. A downshift here—combined with low savings—sets up downside consumption asymmetry (people can’t smooth spending as easily).
• Personal savings rate: 3.6% (May 22) → 2.6% (late May/June) in your history, and 2.7% today. That’s the cleanest “macro vulnerability” trend: households are operating with thin buffers.

Housing

• Building permits: 1442k → 1423k (late May onward). That’s mild deterioration—more “cooling” than “collapse,” but it aligns with your watch status.
• Housing starts: your history shows 1465k through June, while today’s reading is 1239k (WARNING)—implying a material downshift since the early-summer level. If that persists in official releases, housing will become a bigger score driver.

Financial conditions & credit

• NFCI: -0.52 to -0.49 in early June (still loose). Conditions loosened less than before but remain supportive.
• HY spreads: broadly high-200s bps in the history window with one brief spike; today around 271 bps is still complacent/benign.

Markets & valuations (risk concentration)

• S&P 500: in the provided history, ~7446 (May 22) → ~7267 (Jun 11), while today is 7786—a strong rebound since early summer.
• Valuation ratios (e.g., NASDAQ/GDP in DANGER): your market family continues to flag late-cycle excess even while credit remains calm—this is a classic setup where recession risk is not immediate, but drawdown risk can be high if growth disappoints.

Stock Screener Signals

Today’s flagged list is dominated by “value dividend” profiles (ARCC, AIG, BBY, FNF, HMC, T, LTM, BCE) with a smaller sleeve of “oversold growth” (CHTR, TLK). The macro read-through: investors (or your factor model) are leaning toward cash-flow and balance-sheet comfort rather than paying up for high-duration growth.

Two caveats stand out:

• Several yields shown (e.g., ARCC 1002%, BBY 654%) are not economically plausible as sustainable dividend yields and likely reflect data mapping/special distributions/quote anomalies. Even if the absolute yields are distorted, the style signal still matters: the screen is pointing to defensive carry + low P/E.
• CHTR (RSI 28) and TLK (RSI 30) indicate genuine oversold/mean-reversion positioning. Oversold growth flags often appear when the market is jittery about demand (consumer and advertising) or financing conditions—consistent with a “moderate risk” macro tape.

Net: the screen supports a barbell—defensive income/value plus selective oversold rebound candidates—consistent with your overall score holding steady rather than trending risk-on.

Latest Economic Developments

• Fed minutes (released Aug 19, 2026): Multiple summaries report that many officials saw a path where rates might need to rise if inflation fails to cool, reinforcing the Fed’s “conditional hawk” posture. (apnews.com)
  Macro implication: policy is not a guaranteed cushion if growth softens; the Fed is still guarding against inflation re-acceleration.

• Fed decision context (July 29, 2026): The Fed kept the target range at 3.50%–3.75%. (federalreserve.gov)
  Macro implication: holding is supportive versus hiking, but minutes suggest the hiking option remains on the table, which increases downside risk if demand is already cooling.

• Consumer demand data: Retail sales fell 0.6% in July 2026 (and ex-autos/gas also weakened in the cited coverage). (apnews.com)
  Macro implication: this aligns with your “fragile household” thesis: weak sentiment + low savings can translate into abrupt spending pauses.

• Labor market high frequency: A recent report (Aug 13, 2026) noted 209,000 new claims and framed layoffs as still low overall. (apnews.com)
  Macro implication: recession calls typically require claims to break trend for weeks, not days; we’re not there.

• Leading indicators: The Conference Board reported LEI down 0.2% in June 2026 after a +0.1% May gain, with the next LEI release scheduled Aug 20, 2026 at 10:00 a.m. (conference-board.org)
  Macro implication: forward momentum is softening, but not collapsing.

Near-Term Outlook (Next 30 Days)

Base case for the next month: moderate risk, slow growth, high sensitivity to a labor-market inflection.

Key catalysts:

• Weekly initial claims (every Thursday): the trigger is not one print—it’s a multi-week breakout above the ~200–230k regime toward 240k+, then 260k+.
• Conference Board LEI release (Aug 20, 2026): watch whether the LEI resumes declines or stabilizes after June’s -0.2%. (conference-board.org)
• Inflation/PCE narrative: the minutes’ hawkish conditionality means any upside inflation surprise could revive hike pricing and tighten financial conditions quickly.
• Credit spreads: as long as HY OAS stays roughly sub-350 bps, recession risk should remain capped; a sustained widening would be an early warning.

What would move the score?

• MODERATE → ELEVATED: claims trend higher + temp help continues falling + retail spending remains negative.
• MODERATE → LOW/MODERATE (downshift): claims stay contained + LEI stabilizes + Fed rhetoric softens toward patience/easing.

Long-Term Outlook (3-6 Months)

The 3–6 month picture is best described as late-cycle resilience with thinning buffers.

• Why recession is not the base case: labor remains intact, credit is calm, and financial conditions are loose. Tight spreads and a negative NFCI typically precede continued expansion unless a shock hits.
• Why tail risk remains meaningful: the household sector’s low savings and very weak sentiment create nonlinear downside—small negative shocks can translate into larger spending pullbacks. At the same time, Fed minutes show the reaction function is still inflation-first, meaning policy may not cushion growth immediately if demand cracks. (apnews.com)

Historical parallel (framework-level): many slowdowns that avoid recession keep a similar signature—claims stable, spreads tight, but consumer confidence depressed. The difference-maker is whether labor stays firm long enough for real incomes to rebuild buffers, or whether layoffs begin to “follow” weak sentiment.

What to Watch

Hard thresholds (score-moving):

• Initial claims: sustained move >240k for several weeks (early warning), then >260k (material deterioration).
• HY OAS: sustained widening >350 bps (risk-off credit regime).
• NFCI: move from negative toward 0 or positive (tightening conditions).
• Housing: permits sliding toward the low-1300s and starts staying weak would add persistent drag.

Event calendar (next few weeks):

• Weekly claims (Thursdays)
• Conference Board LEI (Aug 20, 2026, 10:00 a.m.) (conference-board.org)
• Next key inflation reads that feed the Fed’s conditional stance (watch for any upside surprises given the minutes’ tone). (apnews.com)

Sources

No data available for this window.