← Back to scans

Will U.S. nonfarm productivity YoY in any quarter for 2026 be above 3%?

KXNFPROD-27MAR04-T3 · Economics · 2026-04-07
43%
Agent
42%
Market Price
+1.0%
Edge
45%
Confidence
Final Rationale
My independent estimate of 40% was based on sub-question probabilities of 14-18% per quarter for exceeding 3% YoY productivity growth, combined with a correlation adjustment. I PARTIALLY ACCEPT the Devil's Advocate critique on several points: (1) the correlation adjustment from ~50% to 40% was indeed somewhat aggressive — effective draws of ~2.8 rather than 2.5 seems more reasonable, pushing toward 45%; (2) the recession scenario from tariff escalation (perhaps 15-20% probability by mid-2026) is underweighted in sub-question estimates, since historically firms shed labor faster than output declines, mechanically boosting productivity YoY above 3% in recessions; (3) I ACCEPT that TFP deceleration from a single year is noisy and the -3% adjustment was oversized. However, I REJECT the implication that the estimate should go as high as 50% because the current trend of 2.1-2.5% YoY is meaningfully below the 3% threshold, and absent a recession shock or AI-driven acceleration, the base trajectory doesn't reach it. The market at 42% is very close to my adjusted estimate, confirming convergence — I land at 43%, reflecting the upward corrections from the critique while respecting that the fundamental macro trajectory remains below threshold.
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 $t{=}0$.
$n = 53$ follow-ups
1. Planner Opus · Fermi decomposition + tool plan
Tools chosen: kalshi_data fred_data article_search web_search code_execution wikipedia_lookup
Sub-questions (Fermi decomposition)
  1. Will U.S. nonfarm business sector labor productivity YoY growth exceed 3% in Q1 2026 (reported ~May 2026)? (w=0.25)
  2. Will U.S. nonfarm business sector labor productivity YoY growth exceed 3% in Q2 2026 (reported ~August 2026)? (w=0.25)
  3. Will U.S. nonfarm business sector labor productivity YoY growth exceed 3% in Q3 2026 (reported ~November 2026)? (w=0.25)
  4. Will U.S. nonfarm business sector labor productivity YoY growth exceed 3% in Q4 2026 (reported ~February 2027)? (w=0.25)
Planner reasoning
This question resolves YES if ANY of the four 2026 quarters shows nonfarm business sector labor productivity YoY growth above 3%. The key insight is that this is a 'maximum of four draws' problem — the probability of at least one exceeding the threshold is much higher than any single quarter exceeding it. The combination logic should be 'maximum' (equivalent to 1 - product of (1-p) for each quarter). Critical factors: (1) Current trend — recent quarters show YoY around 1.8-2.9%, so 3% is achievable but not certain; (2) AI productivity tailwinds may boost productivity in 2026; (3) Base rates from FRED data — historically, >3% YoY has been uncommon outside of cyclical recoveries and productivity booms; (4) Tariff-related economic uncertainty in 2026 may dampen output more than hours, potentially distorting productivity measures; (5) The Q4 2024/Q1 2025 readings will serve as the base for early 2026 YoY comparisons. The question is different from a simple 'will productivity be high in 2026?' because the four-quarter window significantly increases the probability of hitting the threshold at least once. I should also check if the Kalshi market has already priced this.
2. Research Parallel tool calls · legacy v2
ToolStatusTimeItemsSummary
kalshi_data OK 1.2s -
fred_data OK 1.5s -
article_search OK 0.1s -
web_search OK 43.1s -
code_execution OK 0.0s -
wikipedia_lookup OK 1.1s -
3. Evidence Synthesis Sonnet · 16 items · legacy v2
#StrengthCredibilityDirectionSourceClaimPriced In
1 STRONG - ↓ DOWN web_search BLS revised Q4 2025 nonfarm business productivity YoY to 2.5%, with the quarterly rate revised downward to 1.8% (from preliminary 2.8%) due to a 1.1pp downward revision to output. Yes
2 STRONG - ↓ DOWN web_search Annual average nonfarm productivity grew 2.1% in 2025, and TFP grew only 0.8% in 2025 — a step down from 1.5% in 2024 — suggesting decelerating productivity trend. Yes
3 STRONG - NEUTRAL web_search The preliminary Q1 2026 productivity report is scheduled for May 7, 2026, meaning Q1 2026 YoY data is not yet available. Yes
4 MODERATE - ↑ UP kalshi_data Kalshi market KXNFPROD-27MAR04-T3 prices the probability of nonfarm productivity above 3% at 42%, down 14pp over the past 30 days, with a range of 36-77% over 19 data points. Yes
5 MODERATE - ↓ DOWN code_execution Historical base rate analysis (last ~24 quarters): only 4/24 quarters (16.7%) showed YoY productivity >3%; excluding pandemic extremes, 2/22 (9.1%). Mean YoY was 1.98%, stdev 2.45%. No
6 MODERATE - ↓ DOWN code_execution Based on historical base rates, the probability of at least one of four 2026 quarters exceeding 3% YoY is approximately 31.7% (excluding pandemic outliers). No
7 STRONG - NEUTRAL fred_data FRED OPHNFB index was 119.338 as of 2025-10-01, providing a base level for computing YoY comparisons through early 2026. Yes
8 STRONG - ↓ DOWN web_search Q4 2025 YoY productivity of 2.5% provides the recent trend baseline; to exceed 3% YoY in Q2 2026, output growth must accelerate relative to hours relative to Q2 2025 base. Yes
9 STRONG - ↓ DOWN web_search Manufacturing productivity decreased 2.5% in Q4 2025 with unit labor costs rising 9.1%, indicating weakness in goods-producing sectors that could weigh on aggregate productivity. Yes
10 MODERATE - NEUTRAL kalshi_data The same Kalshi market prices the overall annual 2026 productivity-above-3% event at 42%, down from a 30-day high near 56%, suggesting market participants have revised expectations downward. Yes
11 MODERATE - ↓ DOWN web_search TFP deceleration from 1.5% in 2024 to 0.8% in 2025 suggests AI and technology productivity tailwinds have not yet materially accelerated broad-based productivity growth. Yes
12 STRONG - ↓ DOWN web_search Current business cycle (since Q4 2019) productivity growth rate is 2.1% annualized — above prior cycle (1.5%) but below the 3% threshold, suggesting structural productivity is well below 3%. Yes
13 STRONG - ↓ DOWN web_search Unit labor costs rose 4.4% in Q4 2025 and 2.4% over the last four quarters, with compensation growing faster than productivity — suggesting labor market dynamics are not productivity-enhancing. Yes
14 MODERATE - ↓ DOWN kalshi_data Kalshi market for this question has seen significant 30-day price decline (-14pp), suggesting new information (likely the Q4 2025 revised data) moved market expectations downward. Yes
15 STRONG - ↓ DOWN web_search Long-term productivity growth since Q1 1947 is 2.2% per year, and the current cycle rate is 2.1% — both are significantly below the 3% YoY threshold, making sustained above-3% unlikely without a structural shift. Yes
16 MODERATE - ↓ DOWN code_execution The historical base rate of any single quarter exceeding 3% YoY (excluding pandemic) is approximately 9.1%, implying each quarter has roughly a 9-17% chance of clearing the threshold independently. No
Information Gaps
  • No Q1 2026 actual productivity data yet (scheduled May 7, 2026) — the most directly relevant data point for sq1 is unavailable
  • No Q2, Q3, or Q4 2026 productivity data exists yet — sq2, sq3, sq4 rely entirely on projections
  • No detailed economist forecasts for 2026 quarterly nonfarm productivity growth were retrieved
  • No data on tariff impact on output vs. hours in 2026 — tariffs could distort the productivity calculation by reducing output more than hours
  • The web search result for Query 2 (forecasts/economist views) was truncated — full content on 2026 forecasts was not recovered
  • The Kalshi market ticker (KXNFPROD-27MAR04-T3) appears to represent the full-year 2026 question rather than a specific quarter, creating ambiguity about per-quarter probabilities
  • No data on whether AI-related productivity gains are being measured in nonfarm business sector statistics for 2026
  • Base rate calculation covers only 24 recent quarters — unclear which specific time window was used and whether it is representative of the current productivity regime
Key Uncertainties
  • Tariff-related economic uncertainty in 2026: tariffs could reduce output disproportionately relative to hours, potentially suppressing measured productivity
  • AI/technology productivity: if AI adoption accelerates in 2026, productivity could surprise to the upside — but TFP deceleration in 2025 suggests this hasn't yet materialized
  • Base effects: the Q4 2025 downward revision to output creates a lower base that could mechanically boost some 2026 YoY comparisons
  • Potential labor market softening: if companies reduce hours faster than output due to economic uncertainty, productivity could rise mechanically without genuine efficiency gains
  • Downward revision risk: preliminary productivity estimates are frequently revised (Q4 2025 was revised down 1pp), so even if early 2026 readings initially exceed 3%, revisions could bring them below
  • Whether the Kalshi 42% market price reflects the full-year 'at least one quarter' question or a specific quarter question — this materially affects interpretation of the market signal
Market Context
Price: 0.42 · 7d: +0.00 · 30d: -0.14
Volume: MODERATE-HIGH — 2575 average daily contracts is meaningful for a prediction market · Depth: MEDIUM — price range of 36-77% over 19 days suggests some volatility and moderate depth
Reliability: MEDIUM — the market appears to cover the full-year question (any quarter in 2026 above 3%), not individual quarters; the 30-day decline likely reflects the March 24 BLS Q4 2025 revised data release which showed lower-than-expected productivity
4. Independent Forecaster Opus · No market price · legacy v2 Did NOT see market price
Independent estimate: 40% (45% confidence)
sq1: Will U.S. nonfarm business sector labor productivity YoY growth exceed 3% in Q1 2026 (reported ~May 2026)?
14%
Base rate: ~9-17% of quarters historically show >3% YoY productivity growth (excluding pandemic). The Q4 2025 YoY was 2.5%, and the trend has been decelerating (TFP fell from 1.5% to 0.8%). However, the downward revision to Q4 2025 output creates a slightly lower base, and if Q1 2026 output bounces back while hours stay moderate, a >3% YoY is possible but unlikely. The Q1 2025 base was relatively weak (productivity index around 117.5 based on the trajectory), so the YoY comparison isn't particularly easy. Manufacturing weakness (-2.5% in Q4 2025) weighs down. Overall, starting from a ~12% base rate, the decelerating trend pushes down ~2%, but base effects push up ~2%, netting roughly 12-14%. I'll set this at 14%.
sq2: Will U.S. nonfarm business sector labor productivity YoY growth exceed 3% in Q2 2026 (reported ~August 2026)?
15%
Base rate: ~12% from historical analysis. Q2 2025 had a productivity index around 118-119 range. For Q2 2026 to exceed 3% YoY, the index would need to exceed ~121.5-122.5. The current trajectory of ~2.1% annual growth would put it around 121, falling short. However, there's more uncertainty here given it's further out. Tariff impacts could go either way — if output falls but firms rapidly cut hours, productivity could spike; alternatively, output could fall more than hours. There's also the possibility of a recession-induced labor hoarding reversal that temporarily boosts productivity. The additional uncertainty compared to Q1 pushes this slightly above 12%. I estimate 15%.
sq3: Will U.S. nonfarm business sector labor productivity YoY growth exceed 3% in Q3 2026 (reported ~November 2026)?
17%
Base rate: ~12%. More uncertainty at this time horizon. Key consideration: if there's an economic slowdown in mid-2026 due to tariffs or other factors, companies might shed labor faster than output declines, creating a mechanical productivity boost (this happened in 2009 and 2020). Alternatively, AI adoption could start showing measurable effects by H2 2026. The further out we go, the wider the distribution, which actually increases the probability of exceeding a threshold above the mean. The base effect from Q3 2025 is neutral. I push this up to ~17% accounting for wider uncertainty and potential recessionary productivity dynamics.
sq4: Will U.S. nonfarm business sector labor productivity YoY growth exceed 3% in Q4 2026 (reported ~February 2027)?
18%
Base rate: ~12%. The Q4 2025 base is the weakest recent comparison point — the index was 119.338 but with output revised down. To exceed 3% YoY, Q4 2026 would need to be above ~122.9. At the trend rate of ~2.1%, it would be around 121.8, falling short. However, the weak Q4 2025 base is favorable. Maximum uncertainty at this horizon — recession dynamics, AI effects, tariff adjustments all could materially impact either direction. If there's a recession that begins in mid-2026, the labor shedding phase could easily produce a >3% YoY reading by Q4 2026. I estimate 18%.
Show full reasoning trace
base rate: Historical base rate of nonfarm productivity YoY exceeding 3% in any given quarter is approximately 9-17% (depending on whether pandemic quarters are included). Over the last 24 quarters, approximately 4/24 (16.7%) exceeded 3% including pandemic extremes, 2/22 (9.1%) excluding them. The long-run average productivity growth is 2.2%, and the current cycle rate is 2.1% — both well below 3%. The most recent reading (Q4 2025) was 2.5% YoY.
evidence updates: Key downward pressures: (1) TFP deceleration from 1.5% to 0.8% in 2025 suggests no AI-driven structural shift yet (-3%); (2) Manufacturing productivity declining and ULC rising (+4.4% in Q4 2025) suggests cost pressures rather than efficiency gains (-2%); (3) Q4 2025 output revision down by 1.1pp (-2%). Key upward pressures: (1) The lower Q4 2025 base creates slightly more favorable YoY comparisons for later 2026 (+2%); (2) Potential recession-induced labor shedding could mechanically boost productivity in some quarters (+3%); (3) Four independent-ish draws significantly increase the probability of at least one exceeding the threshold (+15% above single-quarter base). Net effect: starting from a single-quarter base of ~12%, the four-quarter window and various adjustments push the combined probability to approximately 40%.
combination method: Maximum (at least one quarter exceeds 3%) with adjustment for positive correlation between quarters. Raw independent calculation gives ~50%, but strong positive correlation (shared macroeconomic conditions, trend persistence) reduces effective independent draws from 4 to ~2.5, bringing estimate down to ~40%.
final: The probability of U.S. nonfarm productivity YoY exceeding 3% in at least one 2026 quarter is approximately 40%. The current trend (~2.1-2.5% YoY) is below the threshold, and there's no clear catalyst for a structural acceleration. However, the four-quarter window provides multiple chances, and tail scenarios (recession with rapid labor shedding, AI productivity surprise, favorable base effects) provide meaningful upside probability. The key uncertainty is whether tariff-related economic disruption will manifest as a productivity-depressing (output falls, hours sticky) or productivity-boosting (rapid layoffs exceed output decline) dynamic.
5. Devil's Advocate Sonnet · legacy v2 Market price revealed
Math Error Asymmetric Evidence Reasoning Flaw
Challenges
  1. The correlation adjustment from 4 independent draws to ~2.5 effective draws is ad hoc and not well-justified. The forecaster states raw independent calculation gives ~50% but then adjusts to ~40%. Using the actual sub-question probabilities: P(none exceed 3%) = (1-0.14)(1-0.15)(1-0.17)(1-0.18) = 0.86 × 0.85 × 0.83 × 0.82 ≈ 0.497, giving P(at least one) ≈ 0.503 if independent. The correlation adjustment to 0.40 implies very strong positive correlation, but the forecaster doesn't rigorously justify the magnitude of this adjustment. In practice, quarters do share macro conditions, but each quarter also has distinct base effects and idiosyncratic shocks. The effective number of independent draws might be closer to 3 than 2.5, which would yield something closer to 45%.
  2. The sub-question probabilities seem somewhat low given the recession scenario is explicitly acknowledged as plausible. The forecaster notes tariff disruptions and potential recession dynamics, yet Q3 and Q4 probabilities are only 17-18%. In historical recessions (2001, 2008-2009), productivity YoY often spiked above 3% as firms shed labor faster than output declined. If there's even a 15-20% chance of a meaningful recession beginning by mid-2026 (which seems reasonable given tariff escalation), and conditional on recession the probability of >3% YoY productivity in at least one quarter is quite high (perhaps 50-60%), this alone would contribute significantly more than the sub-question estimates suggest.
  3. The forecaster weights the TFP deceleration heavily (-3% adjustment) but TFP is a residual measure that can be noisy year-to-year. A single year's decline from 1.5% to 0.8% could easily reverse, especially if the measurement captures lagged effects. This seems like an oversized adjustment for a single data point.
  4. The evidence direction balance is listed as '1 UP, 12 DOWN, 3 NEUTRAL' but this counting method treats each piece of evidence equally regardless of relevance to the specific question (any quarter exceeding 3%). Several of the 'DOWN' items are about the current level being below 3%, which is already captured in the base rate. Double-counting the same information through multiple evidence items creates an illusion of stronger directional evidence than actually exists.
  5. The forecaster doesn't adequately consider the asymmetric nature of productivity shocks. Productivity can spike much more easily than it can collapse in YoY terms, because firms can rapidly cut hours (denominator) during downturns while output adjusts more slowly. This positive skew in the distribution means the probability of exceeding a threshold above the mean may be higher than a symmetric distribution would suggest.
Suggested adjustment: +5pp
The mathematical combination of sub-question probabilities yields ~50% under independence, and the correlation adjustment to 40% appears too aggressive. Additionally, the recession scenario (which has meaningful probability given tariff escalation) is underweighted in the sub-question estimates despite the forecaster acknowledging it. The asymmetric nature of productivity shocks (easier to spike than collapse) further supports a slightly higher estimate. A modest upward adjustment of ~5 percentage points to 0.45 better reflects the mathematics and the acknowledged tail scenarios.
Market comparison: Forecaster: 40%, Market: 42%. The divergence is minimal (2pp) and well within normal uncertainty bounds. The market price is consistent with the direction of adjustment I'm suggesting (slightly higher than the forecaster's estimate). This is not a case where market divergence drives the adjustment — rather, the mathematical and reasoning issues independently support a slightly higher probability.
6. Aggregator Opus · final calibrated forecast
43% (45% confidence)
My independent estimate of 40% was based on sub-question probabilities of 14-18% per quarter for exceeding 3% YoY productivity growth, combined with a correlation adjustment. I PARTIALLY ACCEPT the Devil's Advocate critique on several points: (1) the correlation adjustment from ~50% to 40% was indeed somewhat aggressive — effective draws of ~2.8 rather than 2.5 seems more reasonable, pushing toward 45%; (2) the recession scenario from tariff escalation (perhaps 15-20% probability by mid-2026) is underweighted in sub-question estimates, since historically firms shed labor faster than output declines, mechanically boosting productivity YoY above 3% in recessions; (3) I ACCEPT that TFP deceleration from a single year is noisy and the -3% adjustment was oversized. However, I REJECT the implication that the estimate should go as high as 50% because the current trend of 2.1-2.5% YoY is meaningfully below the 3% threshold, and absent a recession shock or AI-driven acceleration, the base trajectory doesn't reach it. The market at 42% is very close to my adjusted estimate, confirming convergence — I land at 43%, reflecting the upward corrections from the critique while respecting that the fundamental macro trajectory remains below threshold.