← Back to scans

Will there be a recession in 2027? — Yes

KXRECSSNBER-27 · Economics · 2026-08-12
24%
Agent
26%
Market Price
-2.0%
Edge
63%
Confidence
Volume: 149,360
Spread: 1.0c
Days to resolution: 537
Markets in event: 1
Final Rationale
The live Kalshi price of 27% is the primary anchor, but the resolution criterion is strict — two *consecutive* quarters of negative BEA real GDP within a 5-quarter window, which historically occurs in only ~12-17% of windows (~12% post-1985), so the market already embeds an elevated-hazard premium. Current hard data (positive GDP every quarter since Q1 2025, unemployment easing to 4.1%, positively sloped 10y-3m curve, Fed on hold) shows no contraction signal entering the window, and professional forecasts cluster at 16-30% with Moody's 42% an outlier that is also framed around 2026 rather than the specific technical criterion. The devil's advocate is right that current data gives limited visibility into H2 2027 and that a sharp AI-capex unwind could produce a fast two-quarter dip — but Deloitte's own baseline places the sharper correction in 2028, which cuts the other way and offsets that tail. I therefore stay a touch below the anchor at 24%, acknowledging genuine two-sided uncertainty (a plausible range of 18-30%) rather than claiming an edge on the market's direction of travel.
Price vs. prediction
How the market moved after the agent's call. Solid blue line is the actual market price; dashed line is the agent's static forecast at $t{=}0$.
$n = 22$ follow-ups
Re-scan Context
This market has been scanned before. Previous predictions:
DatePredictedMarket PriceConfidence
2026-07-09 37% 40% 50%
2026-06-29 43% 45% 50%
2026-05-13 37% 42% 40%
1. Planner Opus · Fermi decomposition + tool plan
Tools chosen: kalshi_direct kalshi_related polymarket_related fred claude_news gdelt_news code_execution
Sub-questions (Fermi decomposition)
  1. What is the current Kalshi YES price and price history for KXRECSSNBER-27 (recession in 2027)?
  2. What do the companion Kalshi/Polymarket markets for a 2026 recession imply, and is the 2027 price consistent with that term structure?
  3. What is the historical base rate that a randomly chosen 5-quarter window (Q4 2026–Q4 2027 style) contains two consecutive quarters of negative real GDP growth, using BEA quarterly data since 1948 and since 1985?
  4. What is the current trajectory of US real GDP growth, unemployment, payrolls, and the 10y-3m yield curve spread as of the latest data?
  5. What recession probabilities for 2026–2027 are currently published by professional forecasters (WSJ/Blue Chip/SPF surveys, major bank research, NY Fed yield-curve model)?
  6. What identifiable shocks (tariffs, AI capex slowdown, Fed policy path, fiscal drag, labor-market cracks) are cited as raising or lowering 2027 US recession risk?
Planner reasoning
This is a Kalshi economics market on whether two consecutive quarters of negative real GDP growth occur within Q4 2026–Q4 2027, so the Kalshi price is the primary anchor, supplemented by the sibling 2026 recession market for term-structure consistency. The key empirical inputs are the historical base rate of technical recessions in any given ~5-quarter window, current GDP/labor/yield-curve trajectory, and professional forecasters' 2027 recession probabilities.
2. Research Parallel tool calls
ToolStatusTimeItemsSummary
kalshi_direct OK 0.2s 1 ## This Market's Kalshi Data **Yes** (KXRECSSNBER-27) - Current price (probability): 27.00% - 7-day price change: -12.00% - 30-day price change: -11.00% - Average daily volume: 1347 contracts - Price range: 23.00% - 51.00% - Data points: 90 days
kalshi_related OK 2.3s 2 2 related markets / summaries. series KXRECSSNBER: 0 markets (skipped 2 no-signal) | keyword 'recession': ok | keyword 'GDP growth': ok
polymarket_related OK 2.4s 0 Scanned 100 active Polymarket markets, kept 0 matches. keyword 'recession': 0 markets | keyword 'US recession 2026': 0 markets | keyword 'GDP': 0 markets
fred OK 3.8s 6 Fetched 6 FRED series (lookback=1095d).
claude_news OK 31.6s 18 ## Key Findings: US Recession Probability for 2026/2027 **Survey-based recession probabilities (2026, near-term):** - The probability of a recession in the next 12 months fell to 25% from 33% in April, according to the latest WSJ economist survey, with 2026 real GDP growth forecast revised upward
gdelt_news OK 132.9s 20 GDELT: 20 articles across 3 queries (lookback=45d). 'US recession probability 2027': 10 hits | 'recession forecast 2027 economists': error GDELT rate-limited after retries (429) | 'US GDP contraction quarter': 10 hits
code_execution OK 36.8s 0 **Methodology note:** Quarterly real GDP growth flags (negative-growth quarters) were hardcoded from the listed NBER recession episodes (1948‑2025, ~311 quarters, 37 negative quarters). A "hit" = any 5-quarter rolling window containing ≥2 consecutive negative-growth quarters (a rough proxy for "rece
3. Evidence Brief Sonnet · 6638 chars
# Current state KXRECSSNBER-27 resolves YES if BEA data show two consecutive quarters of negative real GDP growth somewhere within Q4 2026–Q4 2027. As of the latest data (through Q2 2026), real GDP growth has been positive every quarter for over a year (no negative quarters since Q1 2025's -0.6%), so the resolution window hasn't even begun yet — this is a pure forward-looking bet on 2026H2–2027 growth turning negative twice in a row. # Timeline of key events - 2025-01-01: Real GDP growth briefly negative (-0.6%), single quarter, not consecutive (confirmed, BEA/FRED). - 2025 Q2–2026 Q2: GDP growth positive every quarter (0.5%–4.4% range), unemployment drifting up to 4.1-4.5%, payrolls flat/slowly rising (confirmed, FRED). - 2026-07 (WSJ survey): 12-month recession probability falls to 25% from 33% in April; 2026 GDP forecast revised up to 2.1% (reported, WSJ/biggo.com). - 2026-Q2 (Philly Fed SPF): "Anxious index" (odds of GDP decline next quarter) at 25.1%; near-term contraction odds ~17.9%, down from 20.9% (confirmed, Philadelphia Fed). - Undated 2026 (cited, exact date unclear): Yahoo Finance reports Kalshi traders pricing 2026 recession at ~17.5% and 2027 at ~41%, and Rosenberg Research/Deloitte flagging 2027 as a more likely contraction year tied to AI-capex unwind (reported). - Last 90 days (through ~2026-08-11): KXRECSSNBER-27 YES price fell from a peak of 51% to current 27%, a sharp de-risking (confirmed, Kalshi direct). # Event Will there be a recession in 2027 (two consecutive quarters of negative BEA real GDP growth between Q4 2026 and Q4 2027)? # Outcomes to forecast Yes / No # Kalshi market anchor Current YES price: **27%**. Trend: -12% over 7 days, -11% over 30 days; 90-day range 23%–51% (peaked near 51%, has since fallen sharply). Avg daily volume ~1,347 contracts — liquid, actively traded market that has been repricing downward recently (Kalshi direct). # Sub-question answers 1. **Current Kalshi price/history** — YES = 27%, down from a 90-day high of 51%; recent momentum strongly downward (-12%/7d, -11%/30d) (Kalshi direct). 2. **Companion 2026 markets / term structure** — No direct KXRECSSNBER-26 Kalshi data was returned (0 markets found), but a Yahoo Finance article cites Kalshi 2026 recession odds at ~17.5% vs. 2027 at ~41%; that 41% figure looks stale/inconsistent with the current 27% 2027 price — the live 27% supersedes the news citation. Term structure (2026 < 2027) is directionally consistent with delayed-risk narratives even after the recent repricing (Kalshi direct vs. claude_news). 3. **Historical base rate for 5-quarter windows with 2 consecutive negative-growth quarters** — Full sample (1948–2025): ~17.3% of rolling 5-quarter windows qualify; post-1985 (Great Moderation era): ~11.9%. Converting hazard rates: 12%/yr → 14.8%, 20%/yr → 24.3%, 30%/yr → 36% per window (code_execution). 4. **Current GDP/labor/yield curve trajectory** — Real GDP growth: +3.8% (Q1'25… wait, per data) most recently +1.5% (2026Q2 annualized), prior quarters 2.1%, 0.5%, 4.4%, 3.8% — no negative quarters since Q1 2025 (-0.6%). Unemployment ticked up to 4.1% (Jul 2026) from 4.5% (Nov 2025). Payrolls flat/slightly rising (~158.9M, essentially plateaued since late 2025). 10y-3m spread positive and volatile (0.69–0.92 over recent weeks), no inversion signal (FRED). 5. **Professional forecaster probabilities** — WSJ survey: 25% (12-mo, as of Jul 2026, down from 33%). Moody's: ~42% for 2026. RSM: 30% (down from 40%). Bloomberg consensus: ~30%, 2% GDP growth. NY Fed yield-curve model: 16.06% (12-months-ahead, for June 2027). SPF "anxious index": 25.1% chance of GDP decline in Q3 2026 (claude_news). 6. **Shocks raising/lowering 2027 risk** — Raising: AI-capex unwind/bubble risk (Rosenberg, Deloitte, SIFMA cite as top downside risk), fading fiscal stimulus, rising corporate refinancing costs/consumer credit stress, labor-market softening/stagflation risk (Stanford SIEPR). Lowering: AI capex still adding an estimated +140-150bp to GDP growth through 2026-27 (Bridgewater), tariff/trade-policy risks reportedly diminishing (SIFMA), Fed funds steady at 3.63% without further tightening (FRED/DFF). # Key facts (high-confidence, factual) 1. [FRED] Real GDP growth has been positive in every quarter from Q2 2025 through Q2 2026 (0.5%–4.4%), following one negative quarter in Q1 2025 (-0.6%). 2. [FRED] Unemployment rose from 4.3% (Aug 2025) to a peak 4.5% (Nov 2025), easing to 4.1% by Jul 2026. 3. [FRED] 10y-3m Treasury spread positive throughout the past month (0.69-0.92), no inversion. 4. [Kalshi direct] YES price for 2027 recession fell from 51% to 27% over past 90 days. 5. [NY Fed] Yield-curve model 12-month-ahead recession probability: 16.06% (for June 2027). # Cross-market signals - Kalshi related: No dedicated 2026 recession series found in this pull (0 markets); GDP-growth-bucket markets (2035/2036) show low probabilities on higher-growth buckets, consistent with modest growth expectations. - Polymarket: No active recession/GDP markets found (0 matches). - Sportsbook implied: N/A. # Analyst opinions and speculation - Rosenberg Research: expects a "very significant" 2027 recession tied to loss of fiscal stimulus + AI capex fade (speculative, single-analyst view). - Deloitte baseline: AI overinvestment correction concentrated in 2027, with sharper downturn (GDP -1%, unemployment 6.5%) more likely in 2028 than 2027 itself. - Bridgewater: cautions current growth is unusually dependent on narrow AI capex profits — a fragile growth driver. # Directional lean per outcome - **Yes**: Rosenberg/Deloitte AI-capex-unwind narratives, elevated Moody's (42%) 2026 estimate, historical base rate uplift for late-cycle windows (24-36% under elevated hazard), Kalshi's own price having been as high as 51% recently. - **No**: Current hard data (GDP, unemployment, yield curve) shows no imminent contraction; WSJ/RSM/Bloomberg surveys trending down (25-30%); NY Fed's mechanical model at just 16%; Kalshi price has fallen sharply to 27%, reflecting cooling sentiment. # Gaps / unknowns - No confirmed live Kalshi 2026 recession price for direct term-structure comparison. - Unclear why Kalshi 27% diverges from cited 41% figure in older news — likely stale reporting; current price should dominate. - No Q3/Q4 2026 GDP data yet (window hasn't started). # Calibration anchors - Kalshi current YES price: 27% (primary anchor). - Professional survey range: 16% (NY Fed) to 42% (Moody's), median ~25-30%. - Historical base rate for 5Q window with 2 consecutive negative quarters: 12-17% unconditional, up to 24-36% under elevated late-cycle hazard assumptions.
4. Forecasters Opus + GPT · in parallel
claude-opus-5
0.62
Yes 24%
No 76%
The Kalshi anchor is 27% YES and falling fast (-12%/7d, -11%/30d from a 51% peak), and I lean modestly below it because the resolution criterion is strict (two *consecutive* quarters of negative BEA real GDP, which has occurred in only ~12-17% of historical 5-quarter windows, ~12% post-1985). Hard data through Q2 2026 show no contraction signal: positive GDP every quarter since Q1 2025, unemployment easing back to 4.1%, a positively sloped 10y-3m curve (0.69-0.92), and the Fed on hold at 3.63%. Professional forecasts cluster at 16-30% (NY Fed yield-curve model 16%, WSJ 25% down from 33%, RSM 30%), with Moody's 42% the outlier. The main YES case is speculative and single-analyst-driven (Rosenberg/Deloitte AI-capex unwind, with Deloitte itself placing the sharper downturn in 2028), and Bridgewater's fragility point argues the risk is real but not yet visible in data. I therefore settle just under the market at 24%.
gpt-5.5
0.64
Yes 26%
No 74%
Anchoring on the live Kalshi YES price of 27%, the market has sharply repriced down from 51%, which points toward No because active traders are de-risking the 2027 recession scenario. Current hard data also points toward No: GDP has been positive through Q2 2026, unemployment has eased to 4.1%, and the 10y-3m spread is positive, so there is no near-term contraction signal as the resolution window approaches. Professional forecasts cluster around 25–30% with the NY Fed model lower at 16%, while historical 5-quarter base rates of 12–17% argue the market price is already elevated versus unconditional history. I keep some substantial YES probability for delayed-cycle risks such as AI-capex unwind, fading stimulus, refinancing stress, and the fact that elevated-hazard scenarios can justify a mid-20s probability, but I see no evidence strong enough to move above the Kalshi anchor.
5. Devil's Advocate Sonnet · red-team critique
1. Both forecasts converge tightly around the Kalshi anchor (24% and 26%) without seriously grappling with the momentum signal itself—a -12%/7d, -11%/30d drop from a 51% peak could indicate either genuine information update or a temporary overcorrection/liquidity-driven swing, and neither forecaster stress-tests why the market moved so fast or whether it might mean-revert before the Q4 2026–2027 window even begins. 2. Neither forecast adequately addresses the Deloitte point that "sharper downturn... more likely in 2028 than 2027 itself" — this is a meaningful signal that the AI-capex-unwind narrative may be mistimed relative to the 2027 resolution window, which should push probability lower than both landed, not just be noted as a caveat. 3. Both forecasters treat the professional-survey range (16-42%) as clustering around 25-30%, but this glosses over the wide dispersion — Moody's 42% is nearly 3x the NY Fed's 16%, and neither forecast explains why they weight the yield-curve model or WSJ survey more heavily than Moody's, which is a real methodological choice left unjustified. 4. Neither forecast considers that the current data (positive GDP for 5 straight quarters, unemployment easing, no yield curve inversion) primarily describes conditions in 2025-H1 2026, while the actual resolution window (Q4 2026–Q4 2027) is 6-18 months out — late-cycle expansions can turn quickly, and anchoring too heavily on "no contraction signal yet" underweights how little visibility current data provides into H2 2027 conditions specifically. 5. Both forecasts land within 2 points of each other and essentially reproduce the Kalshi price with a small discount, offering little independent signal or diversification of view — given the wide disagreement across professional forecasters (16% to 42%) and the base-rate uncertainty (12% to 36% depending on hazard assumptions), a wider spread of considered outcomes (or explicit sensitivity ranges) would have been more informative than converging near-identically on the market price. 6. Neither forecaster explicitly discusses the risk that the "consecutive quarters" resolution criterion could be triggered by a single sharp AI-capex-driven correction (e.g., a sudden pullback in the "narrow AI capex profits" Bridgewater flags as fragile) — this is a plausible tail scenario that could cause a fast two-quarter dip even from currently benign data, and its likelihood/impact isn't quantified separately from the vaguer "speculative" Rosenberg narrative.
6. Aggregator Opus · final calibrated forecast
24% (63% confidence)
Yes: 24% No: 76%
The live Kalshi price of 27% is the primary anchor, but the resolution criterion is strict — two *consecutive* quarters of negative BEA real GDP within a 5-quarter window, which historically occurs in only ~12-17% of windows (~12% post-1985), so the market already embeds an elevated-hazard premium. Current hard data (positive GDP every quarter since Q1 2025, unemployment easing to 4.1%, positively sloped 10y-3m curve, Fed on hold) shows no contraction signal entering the window, and professional forecasts cluster at 16-30% with Moody's 42% an outlier that is also framed around 2026 rather than the specific technical criterion. The devil's advocate is right that current data gives limited visibility into H2 2027 and that a sharp AI-capex unwind could produce a fast two-quarter dip — but Deloitte's own baseline places the sharper correction in 2028, which cuts the other way and offsets that tail. I therefore stay a touch below the anchor at 24%, acknowledging genuine two-sided uncertainty (a plausible range of 18-30%) rather than claiming an edge on the market's direction of travel.
Pipeline Timing
Total pipeline time: 226.3s
Per-tool research timings shown in the Research section above.