← Research · Replication · Macro and cross-asset

Can you trade bitcoin’s reaction to CPI and jobs reports?

The reaction is real but is over within a minute or two. From December 2021 to September 2026 bitcoin fell after hotter-than-expected CPI and payroll numbers and rose after cooler ones, as a 2026 study reports, but entering two minutes after the release and holding for 15 to 60 minutes did not earn more than its costs, and since 2024 the move after entry has been close to zero.

Partly replicatedBTC and ETH, one-minute, US CPI and payrolls releases, Dec 2021 – Sep 2026Last reviewed

The claim

Source we re-testKroner, T. N., Mohammed, I., & Vega, C. (2026). How Do Cryptocurrencies Price Economic News?. SSRN Working Paper 6447644. doi:10.2139/ssrn.6447644

Kroner, Mohammed and Vega study one-minute crypto prices around US macro releases from 2015 to 2025. Volatility, volume and spreads jump at the release and stay elevated for up to half an hour, and surprises that push yields up or risk appetite down push bitcoin and ether down: crypto reacts like equities, not like a currency or gold. For a trader the question is whether the move lasts long enough to catch.

What we tested

We measured each surprise as the gap between the released figure and the market’s consensus the evening before, for CPI and nonfarm payrolls (with the unemployment rate counted in reverse), December 2021 to September 2026. We checked the sign of bitcoin’s move over 5, 15 and 60 minutes, then traded the continuation: enter two minutes after the release in the direction the surprise implies, exit after 15 or 60 minutes, with a round-trip cost of 0.13% for bitcoin.

Result

The direction holds and is strong for CPI: the correlation between the surprise and bitcoin’s move is about −0.6 at every horizon (a hot print, a lower price), and −0.34 to −0.46 for payrolls. Ether moves the same way, by more. But nearly all of the move happens in the first one to two minutes. The continuation trade made +0.05% to +0.23% gross per release and −0.08% to +0.10% after costs, none of it distinguishable from luck, and about zero gross in 2024–2026, when the CPI reaction itself was much smaller than in 2022–2023.

Correlation between the release surprise and bitcoin’s move after it, Dec 2021 – Sep 2026

Window after releaseCPIPayrolls
5 minutes−0.67−0.39
15 minutes−0.60−0.46
60 minutes−0.58−0.34

Negative: a hotter-than-expected number goes with a lower price. Continuation trade from two minutes after the release: −0.08% to +0.10% per release after costs.

What it means

Macro releases move bitcoin like a risk asset, and the direction follows from the surprise — which every other trader can read too, so the price adjusts within seconds. Anyone who reads the number and then clicks is already late; what is left after the first two minutes does not cover the spread and fees.

What would change the verdict

Acting inside the first minute is a question of execution speed rather than forecasting, and we did not test it. A larger reaction could return if surprises again move rates as much as in 2022, when the CPI response was strongest.

This note re-tests a claim made by the cited authors on our own crypto market data; the summary of their method is ours. Tested · published · last reviewed . Research, not investment advice.