A new report from Coinshares challenges widely-circulated claims that Bitcoin mining consumes as much electricity as the entire nation of Ireland. The report estimates that the Bitcoin network actually consumes approximately 35 terawatt-hours (TWh) per year—a figure 115% lower than estimates from Digiconomist's Bitcoin Energy Consumption Index (BECI).
BECI Under Fire
Recent media coverage was triggered by research from Alex de Vries, founder of Digiconomist, which suggested Bitcoin mining power consumption rivals that of Ireland. However, critics have long argued that BECI uses a flawed bottom-up methodology, assuming a small pool of miners represents the entire community. De Vries himself acknowledged that “the information available is pretty poor quality overall” and called for more scientific discussion.Coinshares' report states bluntly: “Our findings strictly contradict both of these figures and we believe that they rest on incorrect assumptions resulting from inadequate research.”
Half the Power, More Renewable
Christopher Bendiksen, head of Coinshares Research and co-author of the report, noted that many miners object to Digiconomist's data. “They don't make their methodology clear; it appears they have taken a bottom-up approach,” he said.The report finds that Bitcoin mining is largely powered by cheap renewable energy, dominated by hydroelectricity, with only limited seasonal use of coal-based generation in certain parts of China.
Hydropower Dominates
China’s southwestern provinces have massive excess hydro generation capacity, and miners often migrate to these regions during wet seasons. The report concludes: “Contrary to previously reported assumptions, Bitcoin mining is largely driven on cheap renewable energy, dominated by hydro, with the limited permanent use of, and some seasonal migrations to, coal-based generation only representing a small part of the network’s total electricity demand.” This provides a more nuanced picture of the environmental impact of cryptocurrency mining, separating fact from inflated estimates.