← Back to scans

Inflation surge in 2026? — At least 3%

KXLCPIMAXYOY-27-P3 · Economics · 2026-04-06
95%
Agent
98%
Market Price
-3.0%
Edge
78%
Confidence
Final Rationale
With headline CPI at ~2.8% in March 2026 needing only 0.2pp more in any of 9 remaining months, the probability of never breaching 3% is very small. Tariff pass-through alone (historically 0.3-0.8pp over 12-18 months) nearly guarantees a breach, and sticky shelter inflation and core CPI already at 3.1% provide additional upward pressure. The Devil's Advocate correctly identified that my 84% was inconsistent with my own overwhelmingly bullish evidence assessment (9 UP vs 3 DOWN factors), and that sq3 inappropriately diluted the estimate. I remain slightly below the 98% market price because the March CPI hasn't been officially released yet (BLS April 10), and there's a small but real tail risk of energy deflation or unexpected demand destruction preventing any single month from breaching 3%.
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 $t{=}0$.
$n = 29$ follow-ups
1. Planner Opus · Fermi decomposition + tool plan
Tools chosen: kalshi_data fred_data article_search web_search code_execution
Sub-questions (Fermi decomposition)
  1. Is year-over-year CPI inflation currently at or above 3% (as of the most recent available month in 2026)? (w=0.45)
  2. Will CPI inflation remain at or rise above 3% year-over-year in at least one of the remaining months of 2026 (May–December)? (w=0.35)
  3. Do current macroeconomic conditions (tariffs, labor market, housing, energy) support sustained CPI at or above 3% in 2026? (w=0.20)
Planner reasoning
This question resolves YES if CPI YoY hits 3% or above in ANY single month of 2026 — making it a 'maximum' type question where only one month needs to breach the threshold. Key considerations: (1) Current CPI trajectory — if already near or above 3%, resolution is highly likely; (2) Tariff impacts from 2025-2026 trade policy may add persistent upward pressure; (3) Base effects from 2025 matter enormously — if 2025 had low inflation months, 2026 YoY comps become easier to beat; (4) The threshold is 3%, which has historically been breached during supply shocks and commodity price spikes; (5) With ~9 months remaining in the year, even a currently sub-3% reading has meaningful probability of seeing at least one spike. The reference class for 'does US CPI exceed 3% YoY in at least one month in a given year' has been common historically but was unusual in the 2010s. The 2021-2024 period saw persistent elevated inflation. Key risk factors: tariff pass-through timing, energy price volatility, shelter component stickiness, and Fed policy response.
2. Research Parallel tool calls · legacy v2
ToolStatusTimeItemsSummary
kalshi_data OK 1.9s -
fred_data OK 1.2s -
article_search OK 9.4s -
web_search OK 37.0s -
code_execution OK 0.7s -
3. Evidence Synthesis Sonnet · 14 items · legacy v2
#StrengthCredibilityDirectionSourceClaimPriced In
1 MODERATE - ↓ DOWN web_search March 2026 CPI YoY came in at 2.8% headline, while core CPI YoY was 3.1% — headline is below the 3% threshold but core exceeds it. Yes
2 STRONG - ↓ DOWN web_search February 2026 CPI YoY was 2.4%, below the 3% threshold, confirming the most recent two months have not individually breached 3% on the headline measure. Yes
3 MODERATE - ↓ DOWN web_search January 2026 CPI YoY was reported but truncated; the trend from Jan-Feb-Mar 2026 shows 2.x% headline readings, none reaching 3%. Yes
4 WEAK - NEUTRAL fred_data FRED CPIAUCSL YoY change listed as +7.7810 for February 2026 — this appears to be an index-point change, not a percentage, and likely reflects raw index-level differences rather than the standard percent YoY; the BLS-reported YoY % for Feb 2026 was 2.4%. No
5 STRONG - ↑ UP kalshi_data Kalshi market for 'CPI YoY at least 3% in any month of 2026' is priced at 98%, up +31% over 30 days, suggesting the market strongly expects the threshold to be breached at some point in 2026 even though January–March have not. No
6 MODERATE - ↑ UP web_search Core CPI for March 2026 was 3.1% YoY, driven by sticky shelter and services inflation, indicating that tariff and housing pass-through could push headline CPI above 3% in coming months. Yes
7 STRONG - ↑ UP web_search Shelter costs (roughly 1/3 of CPI weighting) rose 0.2% MoM in February and lag real-time rents by up to a year, providing a persistent upward force on CPI throughout 2026. Yes
8 WEAK - ↑ UP code_execution Monte Carlo simulation estimates ~68% probability that at least one of the remaining 9 months of 2026 sees CPI YoY at or above 3%, given the current 2.8% level and historical volatility. No
9 MODERATE - ↑ UP kalshi_data The related PPI YoY > CPI YoY market is priced at 93% (up +73% in 30 days), suggesting producers are absorbing significant cost increases that could be passed through to consumers, elevating future CPI. No
10 STRONG - ↑ UP kalshi_data Kalshi overall market at 98% implies participants expect at least one month of 3%+ CPI in 2026, with the sharp 30-day rise (+31pp) suggesting new information (likely tariff escalation and March CPI core reading) drove conviction. No
11 MODERATE - ↑ UP web_search Services inflation remains entrenched as of March 2026, reinforcing the Fed's cautious approach and suggesting core inflation will stay elevated, supporting eventual headline CPI surpassing 3%. Yes
12 WEAK - NEUTRAL fred_data Energy CPI (CPIENGSL) YoY change was only +1.127 index points as of February 2026, suggesting energy is not currently a major upward driver of inflation — a reversal here could push headline above 3%. Yes
13 MODERATE - ↑ UP fred_data Shelter index (CUSR0000SAH1) YoY change was +12.149 index points as of February 2026, confirming shelter remains the dominant inflationary force and a structural upward pressure throughout 2026. Yes
14 MODERATE - ↑ UP kalshi_data The PPI > CPI market pricing at 93% implies significant producer-level cost pressures (likely tariff-driven) that have not yet fully passed through to consumer prices, indicating pipeline inflation risk. No
Information Gaps
  • No direct data on tariff pass-through magnitude or timing — how much of the 2025-2026 tariff increases have already been absorbed vs. are yet to hit consumer prices is unknown
  • No January 2026 CPI YoY figure explicitly confirmed (data was truncated in the web search)
  • No data on labor market conditions in early 2026 (wage growth, unemployment rate) that would affect services inflation
  • No energy price forecast data for May–December 2026 — energy remains the most volatile CPI component
  • No official BLS data for March 2026 directly confirmed (only from a single secondary source, stockpil.com); BLS release was April 10, 2026
  • Base effects for May–December 2026 vs. May–December 2025 not explicitly analyzed — knowing 2025 monthly CPI levels would sharpen the YoY calculation
  • No data on whether the Fed has changed its rate path in response to current CPI readings, which would affect forward inflation trajectory
  • The FRED YoY changes appear to be index-point differences rather than percentage changes — the actual percentage figures need clarification to avoid misinterpretation
Key Uncertainties
  • Tariff pass-through timing: if 2025-era tariffs begin hitting consumer goods in Q2-Q3 2026, headline CPI could surge above 3% even if core trends remain stable
  • Energy price volatility: a commodity spike (geopolitical shock, OPEC cuts) could easily add 0.5–1.0pp to headline CPI in any given month
  • Whether the March 2026 core CPI reading of 3.1% from the secondary source (stockpil.com) is accurate — if so, it represents the first confirmed sub-component breach of the 3% threshold
  • Base effects: if any month in H2 2025 had unusually low CPI (deflation or near-zero MoM), the YoY comparison for 2026 becomes easier to breach 3%
  • Fed policy reaction: aggressive rate cuts could re-accelerate inflation; aggressive hikes could suppress it — current guidance suggests hold posture
  • Conflict between Kalshi market (98%) and Monte Carlo simulation (68%) — the large gap suggests the market is pricing in tariff/pipeline inflation risk not captured in simple trend extrapolation
Market Context
Price: 0.98 · 7d: +0.02 · 30d: +0.31
Volume: MODERATE — 933 avg daily contracts, sufficient for price discovery · Depth: MEDIUM
Reliability: HIGH — strong volume, extreme price movement over 30 days suggests genuine new information (likely March CPI core reading and tariff escalation) drove the market to near-certainty; the wide historical range (44%–99%) shows this was genuinely uncertain recently
4. Independent Forecaster Opus · No market price · legacy v2 Did NOT see market price
Independent estimate: 84% (62% confidence)
sq1: Is year-over-year CPI inflation currently at or above 3% (as of the most recent available month in 2026)?
12%
Base rate: The question asks specifically about headline CPI YoY. Evidence clearly shows Feb 2026 was 2.4% and March 2026 was 2.8% headline. Neither breaches 3%. Core CPI at 3.1% is notable but the question specifies CPI (headline). The Jan-Mar 2026 data consistently shows headline below 3%. There's a small chance the March figure could be revised upward to 3.0%+ or that the April reading (if released before resolution check) already crossed 3%, but based on available data, headline CPI has not yet hit 3% in 2026. P(YES) = 0.12.
sq2: Will CPI inflation remain at or rise above 3% year-over-year in at least one of the remaining months of 2026 (May–December)?
82%
Base rate: Starting from 2.8% in March 2026, we need only a 0.2pp increase in any single month over the next 9 months. Historical CPI volatility: month-to-month YoY swings of 0.2-0.5pp are common, especially when energy prices move. Key upward pressures: (1) Shelter inflation remains very sticky at high levels - this alone provides persistent upward force. (2) Tariff pass-through from 2025-2026 trade policy has likely not fully materialized - significant tariff increases were enacted and the lag to consumer prices is typically 3-9 months, meaning Q2-Q3 2026 could see meaningful pass-through. (3) Core CPI already at 3.1% suggests underlying inflation momentum. (4) Energy is currently a drag but is the most volatile component - any geopolitical shock or OPEC action could add 0.5-1.0pp to headline. (5) Base effects from 2025 H2 could make YoY comparisons easier. The Monte Carlo estimate of 68% seems conservative because it likely doesn't fully account for tariff pipeline effects or energy volatility asymmetry. The fact that we're already at 2.8% with 9 months remaining and multiple structural upward pressures makes breaching 3% in at least one month quite likely. Pushing up from 68% base by ~10-14% for tariff pass-through risk and energy volatility. P(YES) = 0.82.
sq3: Do current macroeconomic conditions (tariffs, labor market, housing, energy) support sustained CPI at or above 3% in 2026?
70%
Base rate: The question asks about supportive conditions rather than actual outcomes. Evidence: (1) Services inflation is entrenched per March 2026 data - MODERATE UP. (2) Shelter remains the dominant inflationary force with 12+ index point YoY increase - MODERATE UP. (3) Energy is currently neutral/slightly deflationary - this cuts both ways but means any energy shock would be additive. (4) Tariff regime from 2025-2026 represents a structural shift that historically adds 0.3-0.8pp to CPI over 12-18 months. (5) Unknown labor market conditions create uncertainty - if wage growth remains elevated, services inflation stays sticky. The conditions are broadly supportive but not overwhelming - energy is a wildcard and the Fed's posture matters. Without aggressive Fed tightening (which seems unlikely given current hold posture), these conditions support at least occasional breaches of 3%. P(YES) = 0.70.
Show full reasoning trace
base rate: Historically, US CPI YoY exceeded 3% in at least one month in roughly 60-70% of years over the past 50 years. In the 2010s decade, it was rare (maybe 2-3 years out of 10). In 2021-2023, it was consistently above 3%. The current trajectory starting from 2.8% in March 2026 with only 0.2pp needed to breach suggests above-average probability for 2026.
evidence updates: Starting from ~70% base rate. Current headline at 2.8% (only 0.2pp away) pushes UP by ~5%. Tariff pipeline effects not yet fully realized pushes UP by ~5%. Sticky shelter inflation pushes UP by ~3%. Energy currently neutral/deflationary pushes DOWN by ~2%. Core already at 3.1% pushes UP by ~3%. Net adjustment: ~+14%, bringing to ~84%.
combination method: Used logical OR structure: P(at least one month) = 1 - P(no month breaches). Combined probability that neither past months (Jan-Apr) nor future months (May-Dec) breach 3% is estimated at ~16%, giving P(YES) ≈ 84%.
final: With headline CPI at 2.8% in March 2026 and only needing 0.2pp more in any single month over 8+ remaining months, combined with tariff pass-through pipeline, sticky shelter costs, and energy volatility potential, I estimate ~84% probability that headline CPI YoY will reach 3% in at least one month of 2026. The main downside risks are aggressive Fed action (unlikely given hold posture), a significant energy price decline, or faster-than-expected shelter disinflation.
5. Devil's Advocate Sonnet · legacy v2 Market price revealed
Math Error Reasoning Flaw Asymmetric Evidence
Challenges
  1. The forecaster's sub-question decomposition has a structural issue: sq3 ('Do current macroeconomic conditions support sustained CPI at or above 3%?') is not logically independent from sq1 and sq2, yet it appears to be used as a multiplicative factor. The resolution question only requires CPI ≥ 3% in at least ONE month — sustainability is irrelevant. Including sq3 as a separate factor dilutes the probability estimate when the question is purely about a single breach.
  2. The combination method claims P(YES) ≈ 1 - P(no month breaches) = 1 - 0.16 = 0.84, but this doesn't follow from the sub-question probabilities. If sq1 (current months already breached) = 0.12 and sq2 (future months breach) = 0.82, then P(at least one breach) = 1 - (1-0.12)(1-0.82) = 1 - 0.88 × 0.18 = 1 - 0.158 = 0.842. This math actually checks out, but sq3 at 0.70 is then unused or confusingly incorporated. The forecaster should clarify what role sq3 plays — if it's meant to condition sq2, then sq2 should be lower unconditionally.
  3. The forecaster notes March 2026 CPI came from 'a single secondary source (stockpil.com)' and flags this as an information gap, yet uses 2.8% as a firm anchor for all calculations. If the actual March figure is different (e.g., closer to 3.0% or already at 3.0%), this would dramatically change the forecast. The uncertainty around this data point should widen the confidence interval.
  4. The forecaster's Monte Carlo estimate of 68% is described as 'conservative' and then adjusted upward by ~14% for tariff effects and energy volatility. However, the Monte Carlo simulation presumably already incorporates historical volatility patterns. Adding an additional 14% for factors that would already be partially captured in historical variance distributions represents potential double-counting.
  5. The forecaster acknowledges the Kalshi market is at 98% but arrives at 84% — a 14pp gap. The market has moved +31% in 30 days on MODERATE volume (933 daily contracts), suggesting genuine information flow. The forecaster's own evidence strongly supports the higher probability: headline at 2.8% needing only 0.2pp, core already at 3.1%, 9 months remaining, sticky shelter, and tariff pipeline. The forecaster's 84% seems inconsistent with their own evidence assessment, which overwhelmingly points upward (9 UP vs 3 DOWN signals).
  6. The sq1 probability of 0.12 may be too low. The forecaster notes March 2026 CPI was 2.8% from a secondary source, but the BLS release date was April 10, 2026 — 4 days after the current date of April 6. This means the March figure hasn't been officially released yet. The forecaster treats this unconfirmed figure as definitive while simultaneously flagging it as potentially unreliable in the information gaps section.
  7. The forecaster underweights the tariff channel. With tariffs described as a 'structural shift that historically adds 0.3-0.8pp to CPI over 12-18 months,' and headline already at 2.8%, even the low end of tariff pass-through (0.3pp) would push headline above 3%. The probability of tariff pass-through being zero or negative is very small, making a breach of 3% nearly certain conditional on tariff effects materializing at any magnitude.
Suggested adjustment: +10pp
The forecaster's own evidence overwhelmingly supports a higher probability. With headline CPI at 2.8% and only 0.2pp needed across 9 remaining months, combined with tariff pass-through (0.3-0.8pp historically), sticky shelter costs, and core CPI already above 3%, the probability of NOT breaching 3% in any single month is extremely low. The sq3 sub-question dilutes the estimate without logical justification for the resolution criteria. The evidence balance (9 UP vs 3 DOWN) and the structural arguments about tariffs alone justify moving closer to 94%.
Market comparison: Forecaster: 84%, Market: 98%. The 14pp divergence appears largely unjustified by the forecaster's own reasoning. The forecaster's evidence strongly supports a higher probability — they identify 9 upward factors vs 3 downward, note that only 0.2pp is needed with 9 months remaining, and acknowledge tariff pipeline effects that alone could push CPI above 3%. The market's 98% may even be slightly generous (energy deflation or unexpected shelter disinflation could prevent breach), but 84% significantly underweights the forecaster's own bullish evidence assessment.
6. Aggregator Opus · final calibrated forecast
95% (78% confidence)
With headline CPI at ~2.8% in March 2026 needing only 0.2pp more in any of 9 remaining months, the probability of never breaching 3% is very small. Tariff pass-through alone (historically 0.3-0.8pp over 12-18 months) nearly guarantees a breach, and sticky shelter inflation and core CPI already at 3.1% provide additional upward pressure. The Devil's Advocate correctly identified that my 84% was inconsistent with my own overwhelmingly bullish evidence assessment (9 UP vs 3 DOWN factors), and that sq3 inappropriately diluted the estimate. I remain slightly below the 98% market price because the March CPI hasn't been officially released yet (BLS April 10), and there's a small but real tail risk of energy deflation or unexpected demand destruction preventing any single month from breaching 3%.