Recession Risk 34/100 — August 19, 2026

Recession Risk Score: 34/100 — MODERATE (-4 vs 30 days ago)

Today’s Recession Risk Score is 34/100 (MODERATE), and it has fallen by 4 points over the last 30 days (from 38 on 2026-07-20 to 34 on 2026-08-19). The signal mix still argues against a recession inside the next ~90 days because the labor-market break that typically precedes imminent recessions is not present: Sahm Rule -0.03 and initial claims ~209K remain firmly in “healthy” territory. Credit is also refusing to validate the consumer-side stress story, with high-yield OAS near ~271 bps—a level consistent with expansionary financial conditions, not a default cycle. The core tension remains: early leading edges are flashing (temporary help, freight, ultra-low savings), but they have not yet cascaded into broad layoffs, tightening credit, or a negative growth print.

---

Score Trend — Last 30 Days

The last 30 days show a net decline from 38 → 34 (Δ -4), with a range of 34 to 44 and an average of 37. The profile is best described as mean-reverting with repeated “risk flares” that fail to persist—spikes into the low-40s get sold quickly back into the mid-30s.

The last 10 readings capture that choppiness: 42 (Aug 10) → 38 (Aug 11) → 34 (Aug 12–13) → 38 (Aug 14) → 37–38 (Aug 15–16) → 34 (Aug 17) → 38 (Aug 18) → 34 (Aug 19). That pattern implies fragile confidence (small shocks still lift risk) but also strong stabilizers (labor + credit + equities) that keep pulling recession odds back down.

Importantly, the score’s floor has been tested repeatedly at 34 (the 30-day minimum) without breaking lower—suggesting that while recession risk has eased versus a month ago, it is not collapsing. In other words: the system is not pricing an imminent downturn, but it is not comfortable enough to move into “LOW” risk either.

---

Key Drivers

1. Labor trigger remains “cold” (Sahm Rule not close)

   • Sahm Rule: -0.03 (SAFE) vs the classic 0.50 recession trigger.
   • This is the single most important near-term recession “gate.” If it’s not rising quickly, the probability of a recession in a 90-day window stays constrained.

2. Jobless claims still historically tight

   • Initial Jobless Claims: 209K (SAFE), still in the “no-wave” regime (your stated healthy band ~200K–230K).
   • This matters because the most reliable short-horizon recession setups are claims-driven: you typically need a sustained move into the mid-250Ks+ plus a rising trend in continuing claims to move from “soft patch” to “hard stop.”

3. Credit spreads confirm calm (no stress pricing)

   • HY OAS: ~271 bps (SAFE), consistent with loose conditions and low perceived default risk.
   • Multiple sources using ICE BofA HY OAS via FRED show readings around 2.71% on Aug 6, 2026. (modigin.com)
   • When recession is near-term, HY spreads usually gap wider first—this isn’t happening.

4. Yield curve normalization reduces near-term recession odds

   • 2s10s: +0.52 (SAFE). A positive term spread materially reduces the probability of a near-term recession relative to an inversion regime.
   • However, note the nuance: 2s30s: 1.06 (WATCH) steepening can sometimes be associated with expectations of future Fed cuts—not bearish by itself, but worth monitoring.

5. Growth is slowing but still positive

   • BEA Q2 2026 real GDP: +1.5% SAAR (advance estimate released July 30, 2026). (bea.gov)
   • Atlanta Fed GDPNow: ~1.8% (your reading), consistent with deceleration, not contraction.

6. Consumer vulnerability is rising (but not yet systemic)

   • Personal Savings Rate: 2.7% (DANGER) + UMich sentiment: 49.5 (DANGER) form the “fragile consumer” cluster.
   • The University of Michigan sentiment data show 49.5 as a recent reading (notably depressed). (sca.isr.umich.edu)
   • This becomes recession-relevant only if it converts into job loss, delinquencies, and credit tightening.

---

Category Breakdown

Using your CATEGORY BREAKDOWN counts:

• Primary Indicators: 4 safe / 4 watch / 1 danger
  Mixed but not recessionary: the primary set is balanced, with labor stress still largely contained.

• Secondary Indicators: 2 safe / 0 watch / 1 danger
  Secondary signals are tilted positive, but the one danger flag suggests “under-the-surface” cyclicals remain soft.

• Housing & Construction: 0 safe / 1 watch / 1 danger
  Housing is deteriorating at the margin (starts/permitting weak), reinforcing the “slow growth” narrative.

• Business Activity: 2 safe / 1 watch / 0 danger
  Business activity is net supportive—notably consistent with your observation that manufacturing is not contracting.

• Consumer Credit Stress: 1 safe / 2 watch / 1 danger
  This is a key watch-zone: delinquency and debt service metrics suggest the consumer is absorbing pressure, but not breaking yet.

• Market Signals: 7 safe / 2 watch / 5 danger
  Markets are sending a split message: index levels/volatility/spreads look benign, while valuation and macro ratios flash “late-cycle/excess.”

• Liquidity: 0 safe / 1 watch / 2 danger
  Liquidity is a soft underbelly today—especially with the ON RRP facility nearly depleted, which can amplify funding sensitivity during shocks.

• Real-Time / High-Frequency: 0 safe / 1 watch / 1 danger
  High-frequency signals remain mixed, with the “danger” component pointing to near-term activity softness (e.g., freight).

---

Biggest Movers

From your BIGGEST MOVERS list (|7-day % change|), with interpretation:

1. Conference Board LEI: +673.3% (7D) — Contradictory / improving

   • The magnitude is clearly a base-effect / sign-flip artifact (moving from a small negative to positive). Directionally, it supports a less imminent recession signal set. (Your model’s LEI reading is positive today; the official June LEI was -0.2% m/m to 99.1.) (conference-board.org)

2. Yield Curve (2s30s): +320.0% (7D) — Ambiguous

   • A fast steepening often reflects rate-cut expectations; it can be either bullish (easier policy) or bearish (growth scare). Given other risk-on indicators (VIX low, spreads tight), this reads more as policy-path repricing than imminent recession.

3. ON RRP Facility: -95.1% (7D) — Confirmatory (higher fragility)

   • Shrinking RRP balances typically imply excess liquidity is being absorbed elsewhere, but at extremes it can reduce the system’s “cash buffer” and increase the importance of bill supply, bank reserves, and money-market plumbing.

4. Freight Transportation Index: -66.7% (7D) — Confirmatory (worsening growth pulse)

   • Freight weakness is a classic early warning for the goods economy. It does not forecast a 90-day recession reliably on its own, but it adds weight to the slowdown case.

5. Sahm Rule: -23.1% (7D) — Contradictory / improving

   • Moving further away from the 0.5 trigger is directly recession-negative (i.e., it reduces near-term recession probability).

---

90-Day Indicator Trends

Your “90-day” history snapshot (as provided) covers a dense window mainly from late May through mid-June for many series, but it’s still useful for mapping direction-of-travel and identifying which signals are drifting versus snapping.

Labor market: cooling, but not breaking

• Initial claims: rose from 211K (2026-05-21) to 225K (2026-06-10)—a mild drift higher, still consistent with a healthy labor market (and consistent with today’s ~209K reading).
• Sahm Rule: eased from 0.13 (late May) to 0.10 (by 2026-06-10) in your history, and is -0.03 today—a continued move away from the recession trigger.
• Quits rate: dipped from 2.0% to 1.9% in early June—consistent with a cooler labor market and reduced worker bargaining power, but not mass-layoff conditions.

Inflection watch: If initial claims push above ~250K and the Sahm Rule climbs sharply (say toward 0.3+ quickly), the risk score would likely re-rate higher fast.

Household resilience: weakening buffers

• Personal savings rate: fell hard from 3.6% (late May) to 2.6% (late May/early June) in your history; 2.7% today remains critically low. This is not a recession trigger by itself, but it creates asymmetry: a small labor shock can transmit quickly into consumption pullbacks.
• Debt service ratio: flat around 11.3%—not explosive, but elevated enough that continued disinflation (or lower rates) would help.

Credit and financial conditions: still expansionary

• HY OAS: tight and stable in the high-200s across your history (271–286 bps), with a brief 320 bps “blip” in your dataset that quickly mean-reverted. Tight spreads remain one of the strongest “no imminent recession” signals.
• Chicago Fed NFCI: moved from about -0.52 toward -0.49—still negative (loose conditions), slightly less loose than late May.

Housing: rolling over

• Housing starts: declined from 1502K to 1465K in late May and stayed there in your history; today you flag 1239K (WARNING), which—if accurate—implies a more meaningful downshift and a potential drag into late 2026.
• Building permits: improved from 1363K to 1442K but then drifted to 1423K—consistent with moderation, not collapse.

Markets: risk-on levels with late-cycle valuation warnings

• S&P 500: rose from ~7433 (May 21) to ~7580 (early June) in your history; today you show 7786—a continued risk-on trend.
• NASDAQ-to-GDP: fell from ~0.85 (early June peak) to ~0.807 (June 10) in your history, but it remains DANGER today (~0.823). That’s a valuation/excess flag—more about future return risk than immediate recession timing.

Bottom line from the 90-day lens: the macro setup looks like cooling growth with strong financial conditions, plus a consumer that is increasingly bufferless. That combination is consistent with soft landing / slowdown, but it becomes recessionary quickly if labor cracks.

---

Stock Screener Signals

Your quant screen is dominated by “value dividend” flags (ARCC, AIG, BBY, FNF, HMC, T, BCE, LTM) and a couple of “oversold growth” names (CHTR, TLK). The positioning signal here is not “panic”; it’s selectivity: the market is hunting for cash-flow durability and yield while still taking opportunistic swings at oversold growth pockets.

Two interpretations stand out:

1. Defensive carry is back in favor (even in an equity-up tape).
   A screen stacked with dividend/value suggests investors want income + margin of safety—a common posture when growth is slowing but recession isn’t priced as imminent. In recession-risk terms, this tends to align with MODERATE: not a bearish stampede, but a preference for sturdier balance sheets.

2. Oversold growth flags imply churn under the surface.
   CHTR (RSI 28) and TLK (RSI 30) point to idiosyncratic drawdowns inside a broader strong index environment. That kind of “internal dispersion” often appears late-cycle: headline indexes hold up while certain cyclicals or leveraged models get repriced.

One red flag: several listed yields are obviously data artifacts (e.g., ARCC “1002%”). Treat the classification (value/dividend vs oversold growth) as the signal—not the literal yield prints.

---

Latest Economic Developments

In the last week (and relevant into today), the data pulse has leaned “slower consumer, still-stable system.”

• Retail sales weakened meaningfully in July.
  The Commerce Department’s July retail sales report (released August 14, 2026) showed headline retail sales down 0.6% m/m, with ex-auto & gas also soft in some summaries—evidence that consumers may be taking a breather after earlier strength. (apnews.com)
  This dovetails with your internal danger signals in savings (2.7%) and sentiment (49.5): spending can cool quickly when buffers are thin.

• LEI confirms a mild leading-edge wobble—not a collapse.
  The Conference Board reported the LEI down 0.2% in June 2026 to 99.1, partially reversing earlier gains and showing a far smaller decline versus late 2025. (conference-board.org)
  This supports the “soft patch” base case: leading indicators are not in a persistent, broad-based decline.

• Fed policy stance remains “hold,” not “tighten.”
  The Fed’s July 29, 2026 FOMC statement held the target range at 3.50%–3.75%. (federalreserve.gov)
  Markets are also keyed to FOMC minutes scheduled for Wednesday, Aug 19 (today). (kiplinger.com)
  For recession risk, the key is whether minutes reinforce a bias to hold (supportive) or re-open tightening risk (negative).

• Markets are not behaving like recession is near-term.
  Recent market recaps show a normal “risk rotation” feel rather than stress—yields steady, equities mixed, and no credit shock headlines. (apnews.com)
  And critically, the HY spread backdrop remains tight around ~271 bps—credit is not pricing a downturn. (modigin.com)

---

Near-Term Outlook (Next 30 Days)

The next month is likely to be dominated by two questions: (1) does consumer softness broaden into labor weakness, and (2) do financial conditions stay loose enough to cushion the slowdown?

Base case (most likely): slow growth, no recession in the next 30 days

• Claims remain contained (sub-240K), the Sahm Rule stays far from trigger, and credit spreads remain tight (<300–325 bps). Risk score likely oscillates low-30s to high-30s, with occasional spikes on data surprises.

Upside risk (risk score down): soft-landing confirmation

• A benign inflation print + stable payrolls + no widening in HY spreads could push the score toward the high-20s/low-30s, especially if housing stabilizes.

Downside risk (risk score up): labor-market discontinuity

• A sustained move in initial claims into ~250K+ and a clear uptrend in continuing claims would likely lift the score quickly into the 40s+, because it would validate the “early warning cluster” (temp help, freight, low savings) as no longer isolated.

---

Long-Term Outlook (3-6 Months)

Over a 3–6 month horizon, the economy looks less like an imminent recession and more like a late-cycle slowdown with asymmetric downside.

• Why recession isn’t “baked in”: financial conditions are still loose (NFCI negative), credit spreads are tight (~271 bps), and the yield curve is positive. Those conditions typically allow weaker pockets (goods, housing) to be absorbed without a full contraction.
• Why risk remains MODERATE: the consumer is operating with thin buffers (savings rate in the danger zone) and confidence is depressed. If labor weakens even modestly, the transmission into consumption could be faster than in prior cycles where households had more excess savings.
• Historical parallel (conceptual): late-cycle phases where equities remain near highs and credit stays tight can persist until claims rise and spreads widen. In those setups, recession risk often re-prices late—meaning the “tell” comes from labor and credit, not from sentiment or temporary help alone.

Net: the 90-day trend supports “slowdown,” but the path dependency is critical. If labor holds, the next 3–6 months can look like a growth scare without recession. If labor breaks, the downside can accelerate quickly because the consumer’s cushion is limited.

---

What to Watch

Hard thresholds that would move the risk score meaningfully:

• Labor market

  • Initial claims: sustained move ≥ 250K (and rising 4-week average)
  • Unemployment rate: continued uptick; any jump that pushes the Sahm Rule sharply higher
  • Sahm Rule: acceleration toward 0.30+ would be an early warning; 0.50 is the classic trigger

• Credit stress / liquidity

  • HY OAS: sustained widening > 350–400 bps (recession risk would likely re-rate quickly)
  • Bank stress proxies: any abrupt tightening in lending standards (SLOOS) or renewed funding market strain
  • Liquidity plumbing: further deterioration in short-term funding conditions as RRP approaches depletion

• Growth / activity

  • Housing: follow-through weakness in starts and permits
  • Freight: stabilization vs continued contraction
  • LEI: watch diffusion—multiple months of broad declines is more recession-relevant than a single down print

• Policy

  • FOMC minutes (Aug 19, 2026): whether the committee signals patience/hold vs renewed tightening bias (kiplinger.com)

---

Sources

No data available for this window.