When to Trust the Crowd Over the Experts.

The Default Assumption Is Wrong

Most founders get their economic outlook in the same way. They listen to their bank’s economist, read City or Wall Street analyst reports, or check what the Bank of England or the Fed has to say. The assumption is that professionals with expensive AI models and proprietary data will always produce better predictions than open platforms where anyone can participate.

A February 2026 research paper from US Federal Reserve economists challenges this directly. The study looked at Kalshi, a regulated betting platform in the US where anyone, from retail investors to professional traders, can place bets on economic outcomes. Will inflation come in above 2.5% this month? Will the Fed cut interest rates at its next meeting? Will unemployment rise? Each bet pays out based on what actually happens. The researchers compared these crowd-sourced predictions against the professional alternatives: economist surveys, interest rate contracts traded by banks and the Fed’s own internal survey of market expectations.

The result: the betting platform’s predictions matched or beat the professionals across nearly every measure. For inflation forecasts, the prediction market was statistically more accurate than the Bloomberg economist survey. For interest rate decisions, the platform’s most-likely-outcome prediction has been correct on the eve of every single rate decision since 2022. A record that neither the professional survey nor the banks’ own trading instruments have achieved.

The study used US data, but the principle is not US-specific. Prediction markets for the Bank of England rate decisions, UK inflation, and UK GDP already exist on platforms like Polymarket. The mechanism, crowds with money at stake outperforming or matching professional forecasters, works the same way regardless of which central bank you are watching.

Why the Crowd Wins (When It Does)

The prediction market is not random luck. It is structural. Understanding why it works tells you when to trust it, and when not to.

It updates constantly, not periodically.

An economist survey is collected once, shortly before a data release. The Bank of England’s own Monetary Policy Report comes out quarterly. A prediction market updates with every trade, which means it absorbs breaking news, MPC member speeches, and surprise data within hours, sometimes minutes. When the MPC voted 5–4 to hold rates at their February 2026 meeting, prediction market prices for the next decision adjusted immediately. Economist forecasts took days to catch up.

People betting money are more honest than people giving opinions.

Survey respondents face no financial penalty for being wrong. Prediction market participants lose real money when their forecast is incorrect. This does not eliminate all biases, the paper finds some evidence that participants slightly overestimate the chances of extreme outcomes, but it imposes a discipline that surveys cannot. When being wrong costs you money, you update your view faster and cling to outdated assumptions less.

It shows you the range of possibilities, not just one number.

A typical economist forecast gives you a single figure: “UK GDP growth will be 1.3%.” A prediction market gives you the full picture: there’s a 30% chance growth exceeds 2%, a 40% chance it’s below 1%, and so on. For a founder deciding whether to hire aggressively or hold cash, knowing there’s a 40% chance of near-zero growth is a fundamentally different planning input than hearing “the consensus is 1.3%.”

Who Should You Actually Listen To?

The takeaway is not that the crowd always beats the experts. It is that different types of economic signals have different strengths, and most founders default to the wrong one for the decision they are actually making.

Prediction markets (Polymarket, Kalshi).

Best for near-term questions with clear outcomes: rate decisions, inflation prints, election results, where speed matters. Weakest for long-term questions where the outcome is years away and fewer people are betting. Example: will the Bank of England cut rates at the next MPC meeting? Polymarket already runs this market and updates in real time.

Economist surveys (Bloomberg, City forecasts).

Best for getting a single, professionally calibrated number when you need one for budgeting or planning. Weakest when the survey was taken days or weeks before your decision and the world has moved on. Example: what do City economists expect UK CPI to be next quarter? The consensus is a useful starting point for a budget.

Central bank statements (BoE Monetary Policy Reports, MPC minutes).

Best for understanding what policymakers intend to do and what they want the market to expect. Weakest when the central bank says one thing but the data pushes them to do another. The BoE’s 5–4 split in February 2026 showed how divided even the MPC can be. Example: the BoE signals a gradual approach. The prediction market says the crowd sees rates stuck. Both signals matter, for different reasons.

Financial market signals (gilt yields, interest rate swaps).

Best for reading longer-term risk and how large institutions are positioning their portfolios. Weakest when prices are distorted by institutional hedging: banks buying insurance against rate rises pushes prices up, making a rate hold look more likely than it actually is. The Fed paper documents this exact problem. Example: gilt yields are sending mixed signals. A prediction market on UK recession probability gives a cleaner answer.

What This Means for Founders

The takeaway is not “replace your CFO with a betting app.” It is that most founders are listening to the wrong sources for the decisions that matter most.

For decisions tied to interest rates (fundraising timing, property, debt financing).

Prediction market probabilities for upcoming BoE decisions are now demonstrably as good as or better than the tools City economists use, at least, that is what the US evidence shows. Check Polymarket’s implied rate outlook before your next board meeting. It updates daily. The forecast your bank gave you was probably formed weeks ago.

For planning under economic uncertainty (hiring speed, budgets, market entry).

Prediction markets show you what the crowd thinks about worst-case scenarios, not just the most likely outcome. If there’s a 40% implied probability of near-zero UK GDP growth, that should change your hiring plan, even if the government’s official growth forecast looks fine.

For making sense of conflicting advice.

When your bank’s economist and the prediction market disagree, the disagreement itself is informative. The Fed paper found that Wall Street’s institutional interest rate instruments were consistently overestimating the likelihood of rates staying high, probably because banks were buying insurance against that scenario, which pushed the price up regardless of what they actually expected. The prediction market, which does not have this hedging distortion, was reading the data more accurately. The same dynamic likely applies to the UK gilt and swap markets.

The Diagnostic

Before making a decision that depends on the economic outlook, ask three questions:

1. 1

   Am I asking about something that will be resolved soon?

   If you are asking “will the BoE cut rates at the next MPC meeting?” or “will UK inflation stay above 3% this year?” prediction markets have been shown, at least in the US, to be as accurate or more accurate than the professionals. Use them as a primary signal, not background noise.

2. 2

   Is the expert forecast stale?

   If the economist survey was conducted before a major data release or geopolitical shift, it is telling you what the world looked like last week. The prediction market has already priced in the new information. The consensus will catch up eventually, but if your decision window is tight, “eventually” is too late.

3. 3

   Do I need to understand the range of outcomes, or just the most likely one?

   If you only need the single most probable outcome, the consensus is fine. If you need to understand the probability of scenarios that would change your plan, a deeper slowdown, a rate hold that lasts longer than expected, inflation staying elevated, prediction markets are the only source that gives you those probabilities in real time.

The crowd is not always right. But the US Federal Reserve’s own researchers have now shown that, for the economic variables that matter most to business planning, the crowd is at least as right as the professionals, and it updates faster, costs nothing to access, and shows you the full picture rather than hiding uncertainty behind a single number. There is no reason to believe the principle stops at the Atlantic.