Recession Risk 38/100 — August 27, 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 (34 on July 28, 2026). The upgrade is not driven by the classic “hard” labor-market recession triggers—those remain largely contained—but by deteriorating cyclicals (freight, temp help, housing) and pockets of consumer strain (delinquencies, low savings). At the same time, financial conditions remain loose and credit spreads remain tight, which typically suppresses near-term recession odds even as late-cycle fragilities build.

Score Trend — Last 30 Days

The last 30 days were defined by range-bound chop with a mild upward drift: Start 34 → End 38 (+4), Avg 37, with a Min of 34 and Max of 44. The distribution matters: we didn’t grind steadily higher; we spiked to 44 at some point in the window, then reverted back toward the high-30s.

The last 10 readings show a clear pattern of mean-reversion (alternating 34/38 prints) that ends with a three-day cluster at 38 (Aug 25–27). That clustering is important: it implies the marginal data flow is no longer “good enough to reset to low-30s,” but also not bad enough (yet) to sustain a push into the mid-40s. In macro terms, this is what late-cycle stabilization looks like: a market and economy that can absorb shocks—until a labor inflection forces repricing.

Key Drivers

Below are the most important drivers of today’s 38/100 print, with the specific readings that matter:

1. Labor triggers still not confirming recession

   • Initial jobless claims: 206K (week ending Aug 15, 2026), still consistent with a healthy layoff backdrop. (content.govdelivery.com)
   • Sahm Rule: -0.03 (SAFE) — firmly below the trigger threshold, meaning unemployment has not accelerated enough to trip the “recession likely underway” condition.

2. Yield curve no longer inverted (near-term risk reduced, but late-cycle signal remains)

   • 2s10s: +47 bps as of Aug 26, 2026, a meaningful positive slope that reduces the probability of an immediate recession call. (yieldcurve.pro)
   • Interpretation: a re-steepening after inversion often occurs late-cycle; it’s not “all-clear,” but it does remove one of the strongest classic recession warnings from the near-term dashboard.

3. Leading indicators improved modestly in July

   • Conference Board LEI: +0.2% m/m in July 2026 to 99.5, after a revised -0.1% in June. (conference-board.org)
   • This reading argues against an imminent, broad-based contraction—especially when paired with tight spreads and buoyant equities.

4. Housing is the clearest macro soft spot

   • Housing starts: 1.239M SAAR in July 2026 and building permits: 1.443M SAAR. (census.gov)
   • Starts are the “here-and-now” construction pulse; permits are the pipeline. The divergence suggests current activity slipped, while the forward pipeline is not collapsing, keeping this as a warning rather than a recession clincher.

5. Credit and financial conditions are still “easy,” delaying downturn dynamics

   • ICE BofA US HY OAS ~2.67% (Aug 14, 2026) / ~2.69% recently—still tight and inconsistent with recession pricing. (fredaccount.stlouisfed.org)
   • Chicago Fed NFCI around -0.56 to -0.57 in mid-to-late August: loose conditions, supportive of risk-taking. (fred.stlouisfed.org)

6. Event risk & macro narrative: Jackson Hole and “AI-cycle” leadership

   • Fed Chair Kevin Warsh is set for a high-stakes Jackson Hole appearance that markets are watching for his inflation/rates stance. (apnews.com)
   • Meanwhile, Nvidia’s Q2 results (a bellwether for capex/AI demand) beat expectations, reinforcing the “growth is uneven but not dead” framing. (apnews.com)

Category Breakdown

• Primary Indicators: 3 safe / 5 watch / 1 danger
  Labor is mostly holding (safe/ watch), but the mix shows creeping fragility—enough to lift the score without flipping the recession call.

• Secondary Indicators: 2 safe / 0 watch / 1 danger
  Secondary signals are not screaming recession, but they’re no longer uniformly supportive—more “late-cycle mixed.”

• Housing & Construction: 0 safe / 1 watch / 1 danger
  Housing is a material drag and remains the cleanest cyclical weakness in the dataset.

• Business Activity: 2 safe / 1 watch / 0 danger
  Business activity is not recessionary, consistent with LEI improvement and resilient corporate conditions.

• Consumer Credit Stress: 1 safe / 2 watch / 1 danger
  This is a slow-burn risk: delinquencies + low savings can turn into demand destruction if hiring cools.

• Market Signals: 6 safe / 3 watch / 5 danger
  The market is simultaneously strong (index levels, low VIX) and stretched (valuation/GDP ratios, commodity fear gauges)—a classic late-cycle contradiction.

• Liquidity: 0 safe / 1 watch / 2 danger
  Liquidity plumbing is less comfortable than risk assets imply; it’s not a recession trigger by itself, but it can amplify shocks.

• Real-Time / High-Frequency: 0 safe / 1 watch / 1 danger
  High-frequency signals are tilting weaker, consistent with freight/temps deterioration.

Biggest Movers

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

1. ON RRP Facility ($405M): +272.7% (7D)
   This is mostly plumbing/technical. Directionally it’s not a clean recession signal, but it reinforces that liquidity regime is shifting (watch shock propagation).

2. NY Fed Recession Probability (0.8%): +33.3% (7D)
   Confirmatory (worsening risk) in rate-model space—probabilities are moving higher even as the yield curve is now positive.

3. Yield Curve (2s10s) (0.47): -29.3% (7D)
   Contradictory (improving near-term recession risk) if the move reflects higher long rates falling or short rates rising less—net result is still a positive curve, but volatility in slope is a late-cycle tell.

4. Housing Starts (1239K): -19.7% (7D)
   Confirmatory (worsening risk) for cyclicals: residential construction tends to lead broader slowdowns.

5. VIX (15.4): +5.2% (7D)
   Mild confirmatory tilt: volatility is rising from complacency, but the level remains low (still “risk-on”).

90-Day Indicator Trends

Your provided 90-day panel shows a macro picture that is more stable than the score’s creep suggests, with weakness concentrated in a few leading micro-cyclicals.

Labor & employment (still the gating factor)

• Initial claims in the history window are still in the low-200Ks (e.g., ~215K to ~229K in late May/June snapshots), consistent with “contained layoffs.”
• Sahm Rule improved versus late May: 0.13 (May 29) → 0.10 (mid-June), and today is -0.03, i.e., moving away from trigger (improving).

Bottom line: your score can rise on cyclicals, but a true recession upgrade generally requires claims upshift + unemployment acceleration—not present yet.

Rates & financial conditions (supportive, but late-cycle)

• 2s10s in the historical slice softened from roughly 0.46–0.47 in late May to 0.29 by June 18, then today is back around 0.47 (as of Aug 26). (yieldcurve.pro)
• NFCI remained loose (negative) in the historical panel and is still negative in August readings. (fred.stlouisfed.org)

Interpretation: the regime is not “tight money causing credit events.” It’s “late-cycle leverage + pockets of strain,” which tends to delay recession timing but can worsen eventual drawdowns.

Housing (clear deterioration)

• In the 90-day history, housing starts were ~1.465M in late May and then printed ~1.177M in mid-June (history), while the latest July official is 1.239M SAAR. (census.gov)
• Permits are ~1.443M SAAR in July, suggesting some stabilization in the pipeline even if activity is soft. (census.gov)

Interpretation: housing is weak enough to keep the score elevated, but not collapsing enough (yet) to force a recession call.

Credit spreads (not recessionary)

• HY OAS oscillated around the high-2s in your history and remains around 2.67% (Aug 14) / ~2.69% recently. (fredaccount.stlouisfed.org)
  Interpretation: credit is not validating the “hard landing” narrative.

Consumer strain & sentiment (bad vibes, but not always action)

• UMich sentiment ~49.5 (danger) sits at crisis-like pessimism, but sentiment can stay depressed for long periods without a recession if labor holds.
• Savings rate very low (~3.0% today; 2.6% in the historical panel), making consumption more fragile if hiring slows.

Stock Screener Signals

Today’s quant flags are overwhelmingly “value dividend” and “oversold growth”—a blend that usually appears when the market is not pricing recession broadly, but is selectively demanding cash flow, balance-sheet quality, and idiosyncratic turnaround setups.

Several flagged names (e.g., ARCC, AIG, BBY, FNF, T) reflect a carry/defensive-income bid: investors are willing to own businesses that can return capital and look cheap on earnings. That aligns with the broader macro setup: financial conditions are loose and spreads are tight, but participants are still subtly rotating toward “paid to wait” positioning.

The oversold growth flags (CHTR, TLK) point to a second theme: mean-reversion hunts in rate-sensitive or balance-sheet-levered growth where positioning may be crowded. In this macro regime, oversold growth can rally sharply if the Fed stays on hold and growth data doesn’t crack—but it’s also the cohort that tends to underperform fast if claims rise or credit spreads widen.

One caution: the displayed yields in the screener output (e.g., triple-digit yields) look mechanically incorrect (likely data or unit issues). Treat the screener as factor-directional (value/oversold) rather than literal yield-level guidance.

Latest Economic Developments

1) Fed narrative: Jackson Hole focus on Warsh’s inflation credibility
Markets are explicitly keyed into Fed Chair Kevin Warsh’s Jackson Hole speech as a moment to clarify his reaction function on inflation and rates. (apnews.com) The significance for recession risk is straightforward: if Warsh leans hawkish (or signals tolerance for higher real rates), the “soft landing supported by easy conditions” can morph into a policy-tightening shock—especially with housing already soft.

2) Policy baseline: July 28–29 meeting held rates at 3.50%–3.75%; minutes released Aug 19
The Fed’s July decision held the target range at 3.5%–3.75%; the minutes confirm a committee still centered on inflation risk management. (federalreserve.gov) For our purposes, this keeps the near-term macro in a wait-and-see mode—supportive for markets, but it increases the importance of incoming labor and inflation prints as catalysts.

3) High yield spreads remain tight (credit not pricing recession)
The ICE BofA US HY OAS around 2.67% in mid-August remains a key “risk-off / risk-on” arbiter: recession risk typically jumps when HY OAS starts trend-widening, not just wobbling. (fredaccount.stlouisfed.org)

4) Housing: July starts at 1.239M SAAR, permits 1.443M SAAR
The July housing report confirms housing as a weak node in the cycle. (census.gov) The question is spillover: will this leak into employment (construction/related manufacturing) and then into claims?

5) Equity leadership remains AI-linked; Nvidia beat reinforces the “growth is uneven” story
Nvidia’s results and outlook were strong enough to push shares higher after-hours, reinforcing investor belief that capex/AI demand remains a tailwind. (apnews.com) This matters because AI-capex strength can offset softness elsewhere—one reason the score stays moderate rather than elevated.

Near-Term Outlook (Next 30 Days)

Base case: moderate risk, range-bound, with recession odds hinging on whether labor finally turns.

Key catalysts in the next month:

• Weekly initial and continuing claims: the cleanest high-frequency recession tell. A sustained move higher (not a one-week blip) would push the score toward 45–60 quickly. The DOL confirmed 206K for the week ending Aug 15, which is still benign. (content.govdelivery.com)
• Jackson Hole (Friday, Aug 28, 2026): Warsh’s speech is a volatility catalyst for rates, the dollar, and risk assets. (apnews.com)
• Financial conditions / HY OAS: watch for spread widening from ~2.7% toward the low-3s and beyond; the direction matters more than the exact level early on. (fredaccount.stlouisfed.org)
• Housing follow-through: August data will tell us whether July was a one-off drop or the start of a lower plateau.

Long-Term Outlook (3-6 Months)

Over a 3–6 month horizon, the economy still looks like a two-speed system:

• Speed 1 (resilient): financial markets, AI-linked capex, and any sector insulated by pricing power and loose conditions.
• Speed 2 (fragile): housing, goods/freight, temp staffing, and lower-income consumers facing higher servicing burdens and low savings cushions.

The structural risk is not that the economy is already in recession; it’s that the cycle is becoming more sensitive to a single labor-market inflection. Historically, once layoffs broaden and unemployment rises persistently, consumer credit stress accelerates, corporate hiring freezes spread, and “moderate” risk can jump to “elevated” in a matter of weeks. Your indicator set is consistent with being one labor turn away from a materially higher score—but that labor turn is not here yet.

The 90-day trajectory reinforces this: housing and select cyclicals are weak, but credit and broad financial conditions are still accommodating. That combination often produces a delayed downturn (or a shallow one), unless policy tightens into the slowdown or a liquidity shock hits the banking system.

What to Watch

High-impact thresholds and events:

• Initial claims: watch for a sustained move above the low-200Ks and acceleration in the 4-week average (directionally, not just one print).
• Sahm Rule: any move toward 0.5 is the red-alert zone; today at -0.03 is comfortably safe.
• HY OAS: if spreads shift from ~2.7% into a persistent widening regime, treat it as confirmation that markets are starting to price recession risk. (fredaccount.stlouisfed.org)
• Jackson Hole (Aug 28, 2026): Warsh guidance that re-prices the September path (hawkish surprise) would tighten conditions quickly. (apnews.com)
• Housing permits vs starts: starts confirm current weakness; permits confirm whether the pipeline is breaking (July permits held up). (census.gov)
• NFCI: a move from comfortably negative toward zero (or positive) would indicate tightening that could transmit into credit and hiring. (fred.stlouisfed.org)

Sources

• conference-board.org
• federalreserve.gov
• federalreserve.gov
• census.gov
• fred.stlouisfed.org
• fred.stlouisfed.org
• yieldcurve.pro
• apnews.com
• apnews.com
• axios.com
• axios.com
• content.govdelivery.com