A new study from University College London reveals that inefficiencies built into Bitcoin’s proof-of-work mining system are wasting staggering amounts of energy — and the problem has grown dramatically over the past decade.
A team of researchers from University College London has put hard numbers on one of cryptocurrency’s least-discussed inefficiencies: the energy burned when two Bitcoin miners solve the same cryptographic puzzle at almost the same moment.
Their findings, published May 26 in PNAS Nexus, reveal that wasted power from these so-called “forks” has surged over the past 10 years, reaching a peak of roughly 16,000 megawatts in 2025 — energy equivalent to about half of the United Kingdom’s total electricity generation.
What Is a Fork, and Why Does It Matter?
To understand the problem, it helps to understand how Bitcoin mining works. Computers on the Bitcoin network race to solve complex cryptographic puzzles in a process called proof-of-work mining. The first machine to crack the puzzle earns the right to add a new block of transactions to the blockchain and collect a fixed reward in newly created Bitcoin. All the other machines that were working on the same puzzle walk away empty-handed.
Occasionally, two miners solve the puzzle nearly simultaneously, each producing a valid block at roughly the same time. This split is called a “fork.” The network eventually settles on one version of the chain — whichever block gets built upon next — while the other block becomes what is known as an orphaned block. Any computing power spent mining on top of that orphaned block is wasted entirely, yielding no reward and no permanent record on the chain.
Lead author Paolo Barucca, an associate professor in the Department of Computer Science at UCL, and his team constructed a mathematical model that captures how fork rates emerge from three interacting factors: the total number of miners competing on the network, the distribution of computing power — or hash rate — among those miners, and the time it takes for a newly solved block to propagate across the network. By fitting that model to real-world Bitcoin data, they were able to estimate just how much electricity gets consumed in dead-end mining efforts.
Power Is Concentrating — and That Changes Everything
The research also sheds light on a structural shift reshaping the Bitcoin ecosystem. Specialized mining hardware, sometimes called ASICs, dramatically outperforms consumer-grade computers, and only well-funded operations can afford to deploy it at scale. That has accelerated a consolidation of mining power: today, just three mining pools are responsible for more than half of all new Bitcoin blocks. This concentration of hash rate affects fork dynamics because larger, better-connected miners can propagate blocks across the network faster, giving them a systematic advantage over smaller competitors.
The model shows that as the industry has consolidated and total network hash rate has climbed, the absolute amount of energy wasted on orphaned blocks has grown sharply, even if the fork rate as a percentage of total blocks has remained relatively small.
Why It Matters for Students and the Planet
For college students interested in fintech, environmental policy, or computer science, this research highlights a tension that sits at the heart of decentralized finance. Bitcoin’s proof-of-work mechanism was designed to be trustless and censorship-resistant, but those properties come with a real environmental price tag — one that extends beyond the energy used to mine successful blocks.
The finding that wasted mining power peaked at around 16,000 MW is striking because it represents electricity consumed with zero productive output. To put that in perspective, the researchers note it is roughly equivalent to half the generating capacity of the entire national grid in the UK. As governments and universities increasingly scrutinize the carbon footprint of digital infrastructure, studies like this one provide the quantitative foundation needed for informed policy debates.
The research also raises questions about whether proof-of-work is the right consensus mechanism for the long term, or whether alternative approaches — such as the proof-of-stake system already adopted by Ethereum — offer a more energy-efficient path forward. Those conversations are directly relevant to students studying blockchain technology, sustainable computing, or energy economics.
What Comes Next
By modeling the relationship between miner heterogeneity, network propagation and wasted computation, the authors have created a tool that could help policymakers, developers and economists better understand the true resource cost of maintaining a decentralized ledger. Future work could extend the framework to other proof-of-work cryptocurrencies, or use it to evaluate how proposed protocol changes might reduce orphaned-block waste.
Source: University College London