Recession Risk 33/100 — May 20, 2026

Recession Risk Score: 33/100 — MODERATE (-14 vs 30 days ago)

Today’s Recession Risk Score is 33/100 (MODERATE), and it has fallen materially (-14 points) over the past 30 days. The key reason is simple: the highest-signal labor-market triggers remain untripped, while financial conditions (outside the long end of rates) still look broadly accommodative. The score is not saying “all clear”—it’s saying the economy is not currently set up for a near-term (next ~90 days) recession unless a shock hits. The main tension is a two-track macro: resilient aggregate labor + markets versus early-cyclical cracks (temp help, freight) and deeply depressed sentiment.

Score Trend — Last 30 Days

From April 20 to May 20, the score moved from 47 → 33 (Δ -14), with a min of 33, max of 47, and average of 40 over 24 samples. The shape is best described as step-down then stabilize: a higher-risk plateau in late April gave way to a lower-risk regime in mid-May, and the last week has mostly mean-reverted around the low-to-mid 30s.

The last 10 readings reinforce that “stabilizing after repricing” profile: 44 (May 11–12) then a sharp drop to 34 (May 13), followed by oscillation between 33–38, ending at 33 today. In recession-risk terms, this pattern usually signals the model is de-emphasizing immediate contraction odds (because labor/credit aren’t breaking) while still flagging late-cycle fragility (because sentiment, goods-cycle, and selected market valuation signals remain hot spots).

Key Drivers

1. Labor triggers remain clearly SAFE (the big offset to fear)

   • Sahm Rule: 0.13 (SAFE)—well below the 0.50 trigger that historically marks a recession signal.
   • Initial jobless claims: 211K (SAFE)—still consistent with a healthy labor market (no layoff wave).
   • SOS indicator: 1.20 (SAFE)—insured unemployment pressure remains low.

2. Yield curve is decisively positive (near-term recession odds typically low)

   • 2s10s: +0.54% (SAFE)—a “normal” curve argues against a recession beginning in the next ~90 days absent a sudden shock.
   • 2s30s: +1.07% (WATCH)—steepening often reflects a market narrative of future easing, which can either be “soft landing” or “cuts because growth breaks.” This is why it stays watch, not safe.

3. Financial conditions are loose, and credit stress is limited (for now)

   • Chicago Fed NFCI: -0.52 (SAFE)—overall conditions are still loose.
   • High-yield OAS: 283 bps (SAFE)—tight spreads argue against imminent default stress.
   • Market backdrop matters here: Reuters coverage in the past 48 hours highlights rising yields tied to inflation/oil/geopolitics, which can tighten conditions quickly if it persists. (investing.com)

4. Growth is slowing but still positive, with mixed forward signals

   • GDP (Q1 advance): +2.0% SAAR (WATCH)—slower but positive.
   • Atlanta Fed GDPNow: 1.8% (WATCH) on today’s tracker—downshift versus earlier highs in your history window, but still expansionary.

5. Early-cyclical deterioration is real (temp help + freight)

   • Temporary help: 2,485K (DANGER)—a classic leading softening signal, often preceding broader labor weakening.
   • Freight index: 1.5 (DANGER)—goods-side demand and transport volumes remain weak.

6. Sentiment is “recessionary,” even if hard data isn’t

   • UMich Consumer Sentiment: 53.3 (DANGER)—crisis-level pessimism raises the odds of a confidence-led demand air pocket, especially if energy prices and geopolitical risk keep pressure on real purchasing power.

Category Breakdown

Using your CATEGORY BREAKDOWN counts:

• Primary Indicators (4 safe / 4 watch / 1 danger): The headline is “labor still holding,” but the cluster of WATCH readings (income, quits, unemployment) says late-cycle cooling is underway even if not yet recessionary.
• Secondary Indicators (2 safe / 0 watch / 1 danger): Mixed but tilted okay; the one danger flag is consistent with the “early cyclical cracks” theme.
• Housing & Construction (1 safe / 0 watch / 1 danger): Housing is not collapsing, but permits remain soft relative to trend—important because housing often turns before the broader economy.
• Business Activity (2 safe / 1 watch / 0 danger): This category is still more “expansion with soft spots” than contraction.
• Consumer Credit Stress (0 safe / 3 watch / 1 danger): This is the sleeper risk: delinquencies and debt service metrics imply less cushion if hiring slows.
• Market Signals (7 safe / 2 watch / 5 danger): Markets are near highs and volatility is low, but valuation/ratio extremes and “fear-metal” signals keep macro fragility elevated.
• Liquidity (0 safe / 1 watch / 2 danger): Liquidity is a latent amplifier—if rates jump or funding strains appear, this is where the score can move quickly.
• Real-Time / High-Frequency (0 safe / 1 watch / 1 danger): Near-term data is mixed; the danger print here aligns with the goods-cycle weakness.

Biggest Movers

From your BIGGEST MOVERS block (|7-day % change|), here’s what changed most and what it means:

1. ON RRP Facility ($13B): -84.7% (7D) — Confirmatory risk (worsening)
   A rapidly depleted RRP balance can be benign (cash moving elsewhere) or can reduce a liquidity buffer. In your model, this keeps liquidity sensitivity elevated.

2. GDP Growth (QoQ annualized) 2.0%: -50.0% (7D) — Confirmatory risk (worsening)
   A sharp downshift in the growth impulse is meaningful even if still positive—especially when paired with weak temp help and freight.

3. NY Fed Recession Probability (2.4%): -30.8% (7D) — Contradictory (improving)
   This is a clean improvement signal: market-implied recession odds (as captured here) are moving lower, consistent with the drop in the headline score.

4. Personal Savings Rate (3.6%): +25.0% (7D) — Contradictory (improving)
   A higher savings rate can rebuild cushion, but the level is still low. If the rise reflects precautionary saving (fear), it can also subtract from consumption—so “improving” with a caveat.

5. VIX (17.8): +14.9% (7D) — Confirmatory risk (worsening)
   Vol is still low, but directionally it suggests markets are paying more attention to rates/geopolitics rather than drifting in a pure risk-on mode.

90-Day Indicator Trends

Your “90-day history” window (as provided) is mostly February–mid-March prints, plus today’s readings. Even within that constraint, several trend messages are clear:

Growth / nowcasting

• GDPNow: shifted from ~3.0% (Feb 19) to ~1.8% (Feb 23 onward) in your recorded history—i.e., a downshift in the growth tracker that persisted through mid-March. Today it’s 1.8% (WATCH), meaning the system is seeing slower-but-positive growth rather than re-acceleration.
• GDP growth (QoQ annualized): your history shows 1.4% through early March, dropping to 0.7% by Mar 14–15 (still positive but weaker). Today’s dashboard shows 2.0% (WATCH), implying the official Q1 print improved versus that late-March internal reading—still consistent with “no imminent contraction,” but not a boom.

Labor-market cooling (without a break)

• Initial claims: tightly rangebound around 206–213K in late February through mid-March; today at 211K—no deterioration trend in the high-frequency layoff signal.
• Unemployment rate: moved from 4.3% to 4.4% in the March portion of your history. Today’s reading is 4.3% (WATCH), which is better than that March uptick, and helps explain why the Sahm Rule remains low.

Rates/curve and risk pricing

• 2s10s: drifted from ~0.61 (Feb 19) down to ~0.55 by mid-March; today at 0.54—basically stable, still positive, still supportive.
• 2s30s: fell from ~1.23–1.28 in late Feb to ~1.12 by mid-March; today 1.07—continued flattening on the long end relative to the front end, consistent with “later easing” expectations.

Financial conditions and credit

• NFCI: stayed around -0.57 to -0.51 in your history; today -0.52—stable and loose.
• HY OAS: mostly high-200s/low-300s bps in late Feb–mid-March; today 283 bps—tight, supportive.

Consumer and early-cycle

• Consumer sentiment: stuck around 56.4 in your history and now 53.3 today—deteriorating at already-depressed levels.
• Temporary help: dropped from ~2480K to ~2447K by early/mid-March; now 2485K today but still marked DANGER on level/trend. Net: persistent weakness in a key leading labor slice.
• Freight index: shifted from +1.3 to -0.5 in early March and stayed there in the recorded history; today it’s flagged DANGER again—persistent goods-cycle softness.

Bottom line from the 90-day lens: labor and credit stayed stable, while growth momentum slowed, and early-cyclical indicators stayed weak. That combination maps exactly to a MODERATE score: not recessionary today, but more fragile than a clean expansion backdrop.

Stock Screener Signals

Your quant flags cluster heavily in “value dividend” (ARCC, AIG, BBY, FNF, HMC, T, BCE, LTM) with a smaller slice of oversold growth (CHTR, TLK). The market message embedded in that mix is not “recession is here”—it’s positioning for carry/quality cashflows while selectively hunting dislocations.

Two nuances matter:

1. Dividend/value dominance = late-cycle caution, not panic.
   When screens consistently surface low P/E + yield names, it often signals the market wants cashflow and defensiveness more than long-duration growth. That aligns with your macro tape: higher long-end yields and inflation sensitivity pressuring duration assets, while the labor market keeps the floor under risk.

2. Oversold growth flags (CHTR RSI 28; TLK RSI 30) suggest selective mean reversion.
   These aren’t broad “buy everything” signals—they’re “some pockets have been hit enough to trade.” In a true recession setup you’d typically see wider credit stress, collapsing cyclicals, and a more systematic risk-off (which your HY OAS and NFCI do not confirm).

One data-quality note you’ll want to fix upstream: several yields are clearly scaled incorrectly (e.g., ARCC “Yield 1002%”). Treat the classification (value/dividend vs oversold growth) as informative, but the absolute yield fields look like a feed/parsing issue.

Latest Economic Developments

Rates and macro narrative (past ~48 hours): the dominant storyline is bond-market strain: Treasury yields have moved higher as investors weigh sticky inflation, elevated oil/geopolitical risk, and growing sensitivity to deficits and term premium. Reuters reporting highlighted the 10-year yield around 4.62% and noted that once long-end yields push through key levels (e.g., 30-year above 5%), markets can feel like they’ve “lost an anchor,” tightening financial conditions even without Fed action. (investing.com)

Equities reacted accordingly: Reuters coverage of May 19 described U.S. equities closing lower with the Nasdaq leading declines as yields hit their highest in more than a year amid inflation concerns and elevated oil, with investors also discussing the possibility the Fed’s next move could be a hike if inflation stays high. (fidelity.com)

Fed communications: Philadelphia Fed President Anna Paulson (per Reuters pickup) said the current stance is appropriate but it’s “healthy” that markets are considering scenarios where rates might need to stay higher for longer—or even rise—given elevated price pressures. This matters for recession risk because it raises the chance of an accidental tightening via markets (higher yields) even if the policy rate holds. (boursorama.com)

Event risk today (May 20): the Fed is scheduled to release FOMC minutes today (for the April meeting), which markets will comb for dissents and the balance of risks. The Fed’s official calendar confirms minutes are due May 20, and market calendars have highlighted this week as a key one for policy narrative under new leadership. (federalreserve.gov)

Near-Term Outlook (Next 30 Days)

The next month is about whether goods-cycle weakness and sentiment infect the labor market—or whether the economy keeps grinding through with claims low and spreads tight.

Base case (most likely): risk score stays in the low-to-mid 30s (MODERATE), fluctuating with rates/oil headlines, because:

• Claims remain near ~200–230K.
• HY spreads remain contained.
• The curve stays positive (2s10s > 0).

What could push the score higher quickly (into the 40s–50s):

• A clear labor inflection: sustained rise in initial claims (especially the 4-week average) plus a rise in continuing claims.
• A “rates shock” that persists: long-end yields rising another leg, tightening mortgage/credit conditions, and hitting confidence and capex simultaneously.
• A credit wobble: HY OAS moving decisively wider (e.g., >400 bps) alongside equity drawdowns.

Near-term calendar catalysts:

• FOMC minutes — May 20, 2026 (today) (narrative risk: higher-for-longer / dissents). (federalreserve.gov)
• Housing starts & building permits — May 21, 2026 (8:30am ET) per multiple market calendars (important because permits have been the softer housing signal). (census.gov)
• Initial jobless claims — May 21, 2026 (8:30am ET) (next weekly print; market calendars show expectations near ~210K). (investing.com)
• May LEI release — May 22, 2026 (per your stated watchlist; LEI momentum can shift “soft landing” narratives fast).

Long-Term Outlook (3-6 Months)

Over a 3–6 month horizon, the macro question is whether we’re in a late-cycle slowdown that stabilizes or a late-cycle slowdown that rolls over.

Why the medium-term still deserves caution even with a MODERATE near-term score:

• Early-cycle labor (temp help) is already DANGER. This tends to lead broader payroll softness.
• Freight/goods weakness is persistent. If it doesn’t turn, it drags manufacturing employment and related services.
• Consumer cushion is thin. Low savings + elevated delinquency stress means consumption can turn abruptly if the labor market softens.
• Fiscal/term premium sensitivity is rising. Reuters’ “bond market strain” framing matters because higher long-end rates can do the Fed’s tightening for it, even if the policy rate stays unchanged. (investing.com)

What would improve the 3–6 month outlook (and keep the score in the 20s–30s):

• Temp help stabilizes and quits stop falling (signaling labor confidence isn’t eroding).
• Freight and cyclicals bottom (inventories normalize, goods demand steadies).
• Long-end yields stop repricing higher (term premium calms), allowing housing and interest-sensitive sectors to breathe.

Historically, “recession surprises” tend to happen when labor breaks after a long lag—the economy looks fine until it doesn’t. Your tracker is correctly centered on that: the Sahm Rule is the fast tripwire, and it’s still far from triggered.

What to Watch

Hard thresholds that would move the risk score:

• Sahm Rule: watch for a move toward 0.30+, then 0.50 (trigger).
• Initial claims: a sustained move above ~250K (and rising 4-week average).
• HY OAS: sustained widening above ~400 bps (stress) or 500+ bps (acute stress).
• 2s10s: a rapid collapse toward 0 (or inversion) would be a major regime change.

Key upcoming events/data:

• May 20: FOMC minutes (tone on inflation persistence and rate path). (federalreserve.gov)
• May 21: housing starts & permits; weekly jobless claims. (census.gov)
• May 22: LEI release (momentum check on forward growth).
• June 5: May jobs report (the single most important “confirm/deny” for labor softening).
• June 16–17: next FOMC meeting on the Fed’s calendar (policy reaction function under the current narrative). (federalreserve.gov)

Sources

No data available for this window.