Two-Way Price Discovery Between a Cryptocurrency-Based Prediction Market and Federal Funds Futures: Evidence from FOMC Decisions

Expectations about policy rates play a central role in financial and insurance markets. This paper asks whether Polymarket, an offshore cryptocurrency-collateralized prediction market, contributes independently to price discovery about Federal Open Market Committee (FOMC) interest-rate decisions or whether its prices merely follow the established Federal Funds Futures (ZQ) market. For twenty-nine FOMC meetings between March 2023 and September 2026, Polymarket event-contract prices and CME Federal Funds Futures prices are each reduced to a meeting-level expected basis-point change on a common time grid and compared along three dimensions: long-run co-movement, forecast accuracy, and the direction of price discovery. The two series are cointegrated in a majority of the meetings and in nearly all of the later more liquid sub-sample, where the long-run relationship is close to one-to-one; the forecast tests do not detect a statistically significant accuracy difference between the markets. Lead–lag tests based on pooled Granger causality and a single-equation error-correction model reveal frequency-dependent predictive relationships. Under strict freshness filtering, a wild cluster bootstrap supports Polymarket-to-futures prediction at 10, 30 and 60 min and the reverse direction at 60 min but neither direction at one minute. Because this longer-horizon pattern cannot be cleanly separated from a timing effect in which both markets respond to common public information at different speeds, the results stop short of establishing that Polymarket leads outright. The findings are consistent with Polymarket being a liquidity-dependent complement to Federal Funds Futures, but do not establish an independent contribution to price discovery.