Do Prediction Markets Forecast Cryptocurrency Volatility? Evidence from Kalshi Macro Contracts

TL;DR

This study shows that daily probability changes in Kalshi prediction markets can forecast cryptocurrency volatility through monetary policy, recession risk, and inflation signals, with evidence of predictive power beyond traditional financial instruments.

Contribution

It provides new evidence that prediction market signals on macroeconomic events can forecast cryptocurrency volatility, demonstrating their informational value and regime dependence.

Findings

Kalshi market probabilities predict Bitcoin volatility with high significance.

Recession risk signals from Kalshi are stable out-of-sample predictors.

Inflation signals forecast altcoin volatility with significant out-of-sample gains.

Abstract

Daily probability changes in Kalshi macro prediction markets forecast cryptocurrency realized volatility through two distinct channels. The monetary policy channel, measured by Fed rate repricing on KXFED contracts, predicts Bitcoin volatility in sample with t = 3.63 and p < 0.001 but exhibits regime dependence tied to the 2024-2025 rate-cutting cycle. The recession risk signal from KXRECSSNBER proves more stable out of sample, delivering an MSFE ratio of 0.979 with Clark-West p = 0.020. The inflation channel, measured by CPI repricing on KXCPI contracts, predicts altcoin volatility for Ethereum, Solana, Cardano, and Chainlink with t-statistics ranging from -2.1 to -3.4 and out-of-sample gains for Ethereum at MSFE = 0.959 with p = 0.010 and Solana at p = 0.048. Both the Bitcoin--Fed-dovish and Chainlink--CPI specifications survive Benjamini-Hochberg correction at q = 0.05. Orthogonalization and baseline comparisons against Fed Funds futures, Treasury yields, and the Deribit implied volatility index confirm that these signals carry information not embedded in conventional financial instruments. The sample covers ten Kalshi event series and six cryptocurrency assets over January 2023 to March 2026.