On Feb. 11, 2026, the U.S. jobs report exposed a wide miss by both professional forecasters and crowds on prediction platforms: economists at major banks expected about 68,000 new jobs for the month, a crowd betting on Kalshi put the number near 54,000, and the Bureau of Labor Statistics reported 130,000 jobs added. A recent National Bureau of Economic Research working paper covering five years of Kalshi data finds that thousands of small-stake bettors often forecast some economic indicators as accurately as highly trained Wall Street economists. The result has drawn attention from market economists and investment analysts who say prediction markets may capture dispersed public information that formal models miss.
Key Takeaways
- On Feb. 11, 2026, consensus forecasts from large financial institutions projected roughly 68,000 new U.S. jobs; Kalshi bettors expected about 54,000; actual payrolls rose 130,000. (BLS)
- An NBER working paper analyzing five years of Kalshi data reports that the crowd’s average accuracy matches that of professional forecasters for several macro indicators. (NBER, 2026)
- Prediction markets like Kalshi and Polymarket also show strong signals for Fed decisions and inflation rates, sometimes outperforming professionals on inflation forecasts. (Working paper summary)
- Thousands of small, anonymous traders contribute to market prices, pooling diverse information and beliefs in real time. (Market observations)
- Some institutional economists, including those at Jefferies, have begun to monitor these market signals as a complementary input to traditional models. (Industry response)
- Prediction markets remain limited by liquidity, platform rules, and regulatory constraints, which affect which questions they can price. (Market structure)
Background
Prediction markets are online exchanges where users buy and sell contracts tied to the outcome of future events. Sites such as Kalshi (launched roughly five years prior to 2026) and Polymarket let participants wager on political events, economic releases, and policy moves, creating a continuous price that can be read as a probabilistic forecast. Economists and traders have long debated whether these markets aggregate genuinely useful information or simply reflect noise from casual bettors.
Traditional macroeconomic forecasting relies on statistical models, survey panels of professional forecasters, and judgment from senior economists at banks and investment firms. Those professionals—often Ph.D.-trained and supported by proprietary data—publish consensus estimates for indicators such as payroll growth, inflation, and unemployment. In some cases, however, those methods suffer from model misspecification, lagging data, or correlated blind spots across institutions.
Main Event
The latest payroll report highlighted the divergence. Ahead of the release, top-bank economists coalesced around a modest gain of roughly 68,000 jobs. At the same time, the Kalshi contract price implied market participants expected near 54,000 new jobs. When the Bureau of Labor Statistics released a headline figure of 130,000, both groups missed by large margins and in similar directions.
The NBER working paper examined five years of Kalshi contracts and compared crowd-derived predictions with professional forecasts across multiple indicators, including payrolls, inflation metrics, and Fed policy actions. The authors found that the crowd’s mean predictive error was comparable to that of institutional forecasters for a subset of outcomes. For inflation measures, the crowd’s median error was often smaller than professional panels.
Industry economists took note. Thomas Simons, a U.S. economist at Jefferies, has said he and colleagues began monitoring prediction-market prices after observing them correctly highlight unlikely candidates in political markets and produce robust signals for economic questions. Market participants on Kalshi are mostly anonymous retail accounts placing small stakes, but their aggregated choices create price trajectories that sometimes move ahead of widely circulated forecasts and market-moving announcements.
Analysis & Implications
If prediction markets reliably aggregate dispersed private information, they offer a low-cost, fast-moving complement to institutional forecasts. Crowds can incorporate real-time observations from geographically dispersed participants, alternative data users, and individuals with diverse priors. That breadth can offset model biases that affect professional forecasters who rely on similar data feeds and techniques.
However, the markets are not a panacea. Liquidity constraints mean some contracts receive sparse trading, which can amplify noise. Platform design—such as contract payoff structures, fee schedules, and eligibility rules—also shapes price discovery. Regulatory limits have prevented broader institutional participation in some cases, narrowing the pool of capital and potentially capping informational efficiency.
For central banks and policy analysts, prediction-market signals could provide a supplementary real-time gauge of market expectations for inflation or policy moves. Yet reliance on these signals requires careful vetting: distinguishing durable signals from short-lived bets driven by news headlines, social media, or strategic trading is essential to avoid overreacting to transient price swings.
Comparison & Data
| Source | Pre-release Estimate (Jobs) |
|---|---|
| Bank Economists (consensus) | 68,000 |
| Kalshi (market implied) | 54,000 |
| BLS actual (Feb. 2026) | 130,000 |
The table above isolates the discrepancy on the February 2026 payrolls number. The NBER paper’s multi-year comparison used mean absolute error and root mean square error across similar episodes to judge relative accuracy. While crowds matched professionals on average for certain indicators, results varied by question type, horizon, and market liquidity. That nuance means prediction-market performance should be assessed case by case rather than generalized to all forecasting tasks.
Reactions & Quotes
Several academics and industry figures framed the finding as a prompt to broaden forecasting toolkits rather than to replace established methods. Below are representative comments and the context for each.
“A wide pool of participants often extracts signals traditional models miss,”—a concise paraphrase of an academic assessment of market aggregation.
Jonathan Wright, Johns Hopkins University (economist)
Wright co-authored the working paper and emphasized that aggregation of many independent judgments can produce a reliable forecast, especially when markets attract informed bettors. He cautioned that market design and liquidity remain critical.
“We began watching prediction markets after they signaled an unexpected political outcome; their real-time pricing is useful as a cross-check,”—a paraphrased industry perspective.
Thomas Simons, Jefferies (U.S. economist)
Simons described how market signals prompted closer internal review of scenarios his team had discounted. He stressed such prices serve as one input among many rather than a single authoritative forecast.
“Retail participation has expanded the information set, but anonymity and low stakes require careful interpretation,”—a paraphrase of market-structure concerns voiced by observers.
Independent market analyst (comment)
Analysts observing Kalshi and Polymarket point to both the informational benefits of many small traders and the need to filter noise from genuine signals.
Unconfirmed
- Whether prediction-market traders possess systematic private information (versus coordinated sentiment or social-media-driven activity) remains unresolved in the public record.
- The extent to which markets would outperform professionals if institutional capital and improved liquidity were broadly allowed onto platforms is not empirically settled.
- Claims that the crowd outperforms experts across all macro indicators are not confirmed; performance appears to vary by indicator and market conditions.
Bottom Line
Prediction markets such as Kalshi and Polymarket are increasingly recognized as useful, often complementary signals for short-term economic outcomes. The Feb. 11, 2026 payroll episode—and the NBER working paper covering five years of Kalshi contracts—illustrate that aggregated small-stake bets can rival professional panels for certain questions, particularly when markets attract active trading and diverse participants.
Policymakers and market economists should consider prediction-market prices as one of several inputs, applying scrutiny to liquidity, contract wording, and potential sources of noise. For investors and analysts, the practical lesson is to monitor these markets as a timely cross-check on model-based forecasts while retaining traditional methods for deeper structural assessment.
Sources
- The New York Times — news report on prediction markets and economists (Feb. 11, 2026)
- National Bureau of Economic Research — working paper repository (academic/working paper)
- Kalshi — prediction market platform (platform)
- Polymarket — prediction market platform (platform)
- Jefferies — investment firm (industry)
- Johns Hopkins University, Department of Economics — academic affiliation for cited author (academic)