Recession Risk 34/100 — May 19, 2026

Recession Risk Score: 34/100 — MODERATE (-13 vs 30 days ago)

Today’s Recession Risk Score is 34/100 (MODERATE), down 13 points versus 30 days ago (47 → 34). The signal mix continues to argue against an imminent, broad-based contraction over the next 90 days because the highest-weight labor trigger—the Sahm Rule—remains decisively “off” at 0.13 (vs 0.50 recession threshold). Markets are also not behaving like a recession is around the corner: the curve is re-steepened, spreads are tight, and overall financial conditions remain loose. The offset is clear: leading cyclicals (temp help, freight, sentiment, household buffers) are deteriorating enough to keep risk moderate rather than low.

Score Trend — Last 30 Days

Over the last 30 days (2026-04-19 → 2026-05-19), the score fell from 47 to 34 (Δ -13), with a min of 33, max of 47, and avg of 40 across 24 samples. The pattern is best described as de-risking with choppiness: a steady downshift interrupted by a few sharp day-to-day spikes (notably the 44 readings on May 11–12).

The last 10 readings show a mean-reverting “settle” into the low-to-mid 30s: after a jump to 44, the score quickly reset to 34 on May 13, then oscillated between 33–38, ending 34 today. This kind of profile usually indicates that systemic stress is not compounding, but pockets of cyclical weakness are still generating “risk flare-ups” that prevent a clean slide into a low-risk regime.

Key Drivers

1. Labor recession trigger remains OFF (core bullish input)

   • Sahm Rule: 0.13 (SAFE) vs 0.50 threshold → no labor-market recession confirmation.
   • Initial claims: 211K (SAFE) (week ending May 9; reported May 14) → still consistent with a labor market that is cooling, not breaking.

2. Curve normalization supports “slow-growth, not recession-now”

   • 2s10s: +0.54pp (SAFE) → a positive curve reduces the classic inversion-imminence narrative.
   • The more nuanced read: steepening can be recessionary only when it’s a “bull steepener” (front-end yields falling on expected Fed cuts due to weakness)—not clearly the case in today’s mix.

3. Credit and financial conditions remain inconsistent with near-term contraction

   • Chicago Fed NFCI: -0.52 (SAFE) and trending loosely supportive of risk-taking and credit creation. (recessionpulse.com)
   • ICE BofA US HY OAS: ~280 bps (SAFE), with May 6 around 2.75%—still “tight spreads” behavior, not “stress spreads” behavior. (ycharts.com)

4. Leading-edge cyclical cracks are real (primary bearish offset)

   • Temporary Help Services: 2,485K (DANGER) → temp help is a classic early-cycle rollover signal.
   • Freight Transportation Index: 1.5 (DANGER) → goods economy weakness remains pronounced.

5. Sentiment is crisis-low—risk to spending and hiring

   • UMich (prelim May): 53.3 (DANGER) in your dashboard. (Note: some public reports show a lower preliminary May reading; regardless, the signal is “deep pessimism,” not “normal confidence.”) (sca.isr.umich.edu)

6. Growth is slowing, but not collapsing (late-cycle, asymmetric downside)

   • April payrolls +115K; unemployment 4.3% → deceleration but still expansionary. (bls.gov)
   • Atlanta Fed GDPNow: 1.8% (WATCH) → below-trend but not recession-like in isolation. (atlantafed.org)

Category Breakdown

• Primary Indicators (4 safe / 4 watch / 1 danger): The primary set is mixed, but the key point is labor-trigger safety (Sahm) is dominating the “no imminent recession” message.
• Secondary Indicators (2 safe / 0 watch / 1 danger): Secondary signals are mostly stable; the danger print implies fragility in the next layer down rather than broad breakage.
• Housing & Construction (1 safe / 0 watch / 1 danger): Housing is bifurcated—starts are okay, but permits are weak, suggesting future pipeline softening.
• Business Activity (2 safe / 1 watch / 0 danger): Business activity is holding up, consistent with “slow growth” rather than contraction.
• Consumer Credit Stress (0 safe / 3 watch / 1 danger): This category is quietly important: no “safe” readings means household balance sheets are less able to absorb shocks if unemployment rises.
• Market Signals (7 safe / 2 watch / 5 danger): Markets are giving a two-handed signal: broad indices and volatility look fine, but valuation/ratio extremes and cyclicals (e.g., copper/gold) flash macro caution.
• Liquidity (0 safe / 1 watch / 2 danger): Liquidity is a weak spot, consistent with reduced system buffers (e.g., depleted RRP).
• Real-Time / High-Frequency (0 safe / 1 watch / 1 danger): Real-time data is mixed; the danger print means the next 2–6 weeks matter for confirmation.

Biggest Movers

1. ON RRP Facility ($7B): -84.7% (7D)
   Confirmatory (worsening risk via thinner liquidity buffers). Rapid depletion reduces “shock absorbers” in money markets.

2. GDP Growth (QoQ annualized): -50.0% (7D)
   Confirmatory (worsening risk). Even if noisy, the direction is consistent with late-cycle downshifting.

3. NY Fed Recession Probability (3.0%): -30.8% (7D)
   Contradictory (improving). A sharp drop reinforces the “not imminent” view—though probabilities can lag turning points.

4. Personal Savings Rate (3.6%): +25.0% (7D)
   Contradictory (improving), but context matters. A rise helps, yet the level remains thin versus pre-2020 norms.

5. VIX (18.4): +14.9% (7D)
   Confirmatory (worsening risk at the margin). Volatility rising from a low base suggests markets are starting to price fatter tails, even if not panic.

90-Day Indicator Trends

Your 90-day history window (mostly Feb–Mar 2026 observations shown) tells a consistent story: macro stress is not erupting, but cyclical internals are deteriorating and liquidity/valuation tail risks are building.

Rates/curve (stability with normalization):

• 2s10s hovered around 0.62 → 0.55 from Feb 18 to Mar 16, staying positive throughout that sample. That’s consistent with today’s +0.54 reading: the curve message is stable expansion/slowdown, not imminent recession.
• 2s30s declined from ~1.24–1.28 in late Feb to ~1.12 by Mar 14–16. Today you’re around 1.03 (WATCH)—a continued compression that can matter if it reflects front-end easing expectations rather than long-end inflation premia.

Financial conditions and credit (still supportive):

• NFCI was tightly range-bound around -0.57 to -0.51 across late Feb to mid-March, and it’s still -0.52 in mid-May—conditions remain loose, not tightening into recession.
• HY OAS in the late-Feb to mid-March history moved roughly 286–317 bps—modestly wider at moments of risk, but not trending into stress. Your current ~280 bps looks like re-tightening, a recession-negative signal.

Equities and risk pricing (still “risk-on,” with valuation caveats):

• In the history sample, the S&P 500 fell from ~6881 (Feb 18) to ~6632 (Mar 15), while today your dashboard prints 7409—a major rally since that period.
• VIX rose materially in early March (up into the high 20s in your history sample) but sits 18.4 today—implied volatility is back to complacent-ish levels.

Labor (cooling but not triggering):

• Initial claims stayed around 206–213K across late Feb–mid March; today you’re at 211K—still consistent with low layoff intensity.
• Sahm Rule moved from ~0.30 (watch) in late Feb/early Mar to 0.27 (safe) by Mar 8–16, and now 0.13 (safe)—a clear trend away from recession confirmation.

Consumer/household buffers (structural fragility):

• Savings rate was 3.6% throughout the Feb–Mar sample, briefly showing 4.5% in mid-March in your history feed—today it’s back to 3.6% (WARNING) in your “today” list, implying households remain buffer-thin.
• Credit card delinquency sits around 2.9% in your history and today—steady, but elevated enough to matter if unemployment rises.

Net: the 90-day direction of travel most consistent with your score decline is labor safety + easy financial conditions, while the components preventing a “low risk” score are temp help, freight, sentiment, liquidity depletion, and valuation/tail-risk market indicators.

Stock Screener Signals

Today’s quant flags cluster into two themes: (1) value/dividend defensives and (2) selective oversold growth.

First, the heavy representation of value dividend names (ARCC, AIG, BBY, FNF, HMC, T, BCE) is classic late-cycle positioning—investors leaning toward cash-flow yield and lower multiples rather than paying up for long-duration growth. That aligns with the macro tape implied by your dashboard: moderate recession risk, easy-ish financial conditions, but rising uncertainty around the consumer and cyclicals.

Second, the “oversold growth” flags (e.g., CHTR RSI ~28, TLK RSI ~30) suggest a market that is not broadly risk-off (indices near highs), but where pockets are being punished—often sectors exposed to rate sensitivity, household pressure, or leverage. In recession-risk terms, this is consistent with a regime where the aggregate economy muddles through, but distributional stress increases (some industries/households deteriorate even as top-line GDP remains positive).

A data hygiene note: the listed dividend yields (e.g., ARCC 1002%) are almost certainly artifacts (special dividends, annualization quirks, or data errors). Treat the directional message—“yield/cheapness is being screened in”—as the signal, not the literal yield numbers.

Latest Economic Developments

Leading indicators: The Conference Board reported the US LEI fell 0.6% in March 2026 to 97.3, reversing February’s +0.3%, and explicitly noted the next release is scheduled for Friday, May 22, 2026 (10:00 a.m. ET). (conference-board.org) This is important for the next 30 days: LEI is one of the cleanest “slowdown accelerant” series—another weak print would strengthen the case that temp help/freight weakness is translating into broader activity.

Manufacturing: April’s ISM Manufacturing PMI held at 52.7 (expansion), but the Employment Index fell to 46.4 (deeper contraction). (prnewswire.com) This “output up / hiring down” split fits your current diagnosis: businesses can maintain production via productivity, hours, and inventory control—until demand softens enough that they cut payrolls.

Labor market: The BLS reported April payrolls +115,000 and unemployment 4.3% (Employment Situation released May 8, 2026). (bls.gov) The gain is not recessionary, but the scale is consistent with a labor market that is losing momentum—which matters because your recession call is anchored to the Sahm mechanism (unemployment rate dynamics).

Financial conditions and credit: Chicago Fed NFCI sits at -0.52 as of May 14, 2026, indicating loose conditions. (recessionpulse.com) In parallel, HY OAS levels around 2.75–2.82% in early May remain tight. (ycharts.com) The combined message is that credit markets are not transmitting a recession signal yet.

Rates/market tone: Recent market coverage highlighted rising Treasury yields and equity sensitivity to rate moves (a “bear steepening” dynamic in mid-May). (kiplinger.com) This matters because if the curve steepens due to higher long rates (inflation/term premium) rather than lower short rates (growth scare), recession odds typically do not rise immediately—but financial conditions can tighten for housing and capex if yields stay elevated.

Near-Term Outlook (Next 30 Days)

Base case for the next month: moderate risk with a slight downside skew—i.e., a score range roughly 30–45, with the direction determined by whether labor and credit remain calm as cyclicals crack.

Key catalysts in the next 30 days:

• Conference Board LEI (Apr 2026) on May 22, 2026: a second consecutive weak print would increase confidence that the slowdown is broadening beyond “vibes” and goods/freight. (conference-board.org)
• Jobless claims (weekly): watch for a sustained move above the low-200Ks baseline. Your current level (211K) is fine; what would matter is a persistent upshift (not one noisy print).
• May employment report on June 5, 2026: payrolls and unemployment will directly feed the Sahm trajectory. (bls.gov)

What would push the score lower (less risk):

• LEI stabilizes or rebounds, claims stay near ~200–220K, and credit spreads remain tight.

What would push the score higher (more risk):

• Temp help continues falling and is joined by rising claims/continuing claims; unemployment ticks up enough that Sahm moves materially toward 0.50; HY OAS widens sharply from ~280 bps.

Long-Term Outlook (3-6 Months)

Over a 3–6 month horizon, the macro regime still looks like late-cycle slow growth with asymmetric downside—meaning recession is not the modal outcome, but the “bad outcomes” can arrive quickly if the labor market turns.

Three structural issues keep tail risk alive:

1. Household buffers are thin (low savings; rising delinquency), making consumption more sensitive to job/income shocks.
2. Goods/industrial cyclicals are already soft (freight and temp help signals), which historically often precede services-labor deterioration.
3. Liquidity backstops look thinner (RRP near depleted), increasing the chance that a market or funding disturbance transmits faster than it would with larger system buffers.

Counterweights that reduce recession likelihood:

• Financial conditions are still loose (NFCI negative).
• Credit spreads are tight, implying investors do not see widespread defaults or refinancing stress as imminent.
• The Sahm Rule is far from trigger, and historically recessions rarely begin with that indicator still this low.

The key long-horizon “tell” is whether the curve’s steepening becomes bull steepening (front end falling on growth fears) and is accompanied by credit widening. If you get both, the probability of a recession within 6 months rises materially—even if the next 90 days remain ambiguous.

What to Watch

Concrete thresholds and events:

• Sahm Rule: watch for 0.13 → 0.30 (first warning) and 0.50 (trigger).
• Initial claims: a sustained regime shift above the low-200Ks baseline; acceleration matters more than level.
• HY OAS: sustained widening above ~350–400 bps would be a meaningful “risk-on → risk-off” transition from today’s tight levels.
• NFCI: a move from ~ -0.5 toward 0 would indicate material tightening.
• LEI (May 22, 2026): another negative print would reinforce that the slowdown is broadening.
• June 5, 2026 jobs report: unemployment rate and participation dynamics; not just payroll headline.
• Goods-cycle confirmations: continued declines in temp help and freight without offset from real income gains.

Sources

• conference-board.org
• recessionpulse.com
• ycharts.com
• bls.gov
• atlantafed.org
• prnewswire.com
• sca.isr.umich.edu
• bloomberg.com
• kiplinger.com