Hello Medium, I’m Christopher Bendiksen — I head up research at CoinShares; lately we’ve heard some rather unsubstantiated claims about the energy consumption / climate impact of the bitcoin network repeated through various media channels.
The volume of these claims has gotten too loud to ignore — so, we’ve written a comprehensively researched paper on the current state of Bitcoin mining network.
We intend for this to be the first iteration in a series of many such papers designed to keep the debate honest and — to the highest degree possible — based on facts.
In case you are pressed for time, here are 6 key takeaways:
1. Bitcoin’s Energy Usage Is A Major Component Of Its Network Security Model
Unlike previous centralised network infrastructures where a third party is in total control, in a distributed system, part of the key to keeping the transaction record honest is determining who gets to write to the database.
In the Bitcoin protocol, any entity requesting write privileges (adding pages) to the distributed database (ledger) has to prove an expenditure of energy (work).
Why? They need to prove that it cost them something to contribute to the network; a contribution for which they are subsequently rewarded with bitcoins.
No risk (expenditure), no reward (bitcoin).
The reward being paid in bitcoin ensures the incentives are aligned between maintaining truthfulness (security) of the network and mining (writing to) the network; because acting in bad faith for the network results in diminishing the value of the bitcoin reward (financial loss for the actor).
The end result is that any reasonable attempt to fraudulently amend or append the database requires a CAPEX investment that is larger ($) than the size of the existing (honest) mining network. The larger the on-going energy expenditure and prior CAPEX expenditure, the larger the required investment to attack.
2. Prior Reports On Bitcoin’s Energy Footprint Were Greatly Exaggerated
Our calculations show prior reports overestimated by a factor of nearly two*.
Where our research produced an estimate of 35 TWh in mid-May, prior estimates had that at nearly double — 65 TWh*. We would also like to note that this is simply a criticism of previous methodologies. If the previous estimate had indeed been correct, it would suggest an even higher level of security than what is currently achieved.
3. Hydro Power Actually Appears To Be The Primary Electricity Source Of The Bitcoin Mining Network
Counter to prior reports of the bitcoin network’s carbon footprint (32M Tonnes) — our research: 1) finds no proof of this claim, and 2) in fact identified the primary source of power for the mining network to be hydroelectricity*.
The reason is simple, coal is too expensive. Bitcoin mining is extremely competitive and the cheapest available electricity is most often found at stranded or seasonally overproducing renewable generation sites.
4. ‘Migratory Miners’ Seasonally Move To Wherever Conditions Are Most Suitable (Profitable)
Price pressures on electricity for mining are so strong that some miners migrate with the seasons to capture the cheapest renewables, in the most advantageous climates, often when seasonally variable production outstrips the steadier demand.
A side effect of this is that there is evidence of miners tapping into more remote/stranded/newly built sources of renewables due to the high mobility of bitcoin miners and immobility of generation infrastructure.
In this sense Bitcoin miners use bitcoin as a “synthetic battery” of sorts, converting otherwise unproductive or wasted electricity generation into spendable monetary assets.
5. Annually, Network Hashrate Grows By ~300%, While Chip Efficiency Grows By ~80% And Chip Cost Per Hash Falls by ~50%
Over the last 4.5 years, the network hashrate has grown by approximately 300% annually*.
Over the same time span, chip efficiency, measured in GH/J has increased by an average annual rate of nearly 80% while the chip cost per hash ($/TH/s) has fallen by an annual rate of nearly 50%*.
This means that every year, a miner’s dollar buys them nearly twice as much hashpower as the year before (CAPEX), and that new hashpower draws only a little more than half the electricity (OPEX).
6. We’re Observing Increasing Geographical Distribution Among Miners
In the wake of the unfriendly policies towards Bitcoin miners adopted by the Chinese government we areobserving an increase in miners setting up shop in the Nordic countries, Canada and north-western United States.
Common to all these locations are cool climates, availability of cheap renewable energy, high-speed internet and business-friendly governments.