I think a lot of the outrage about proof-of-work-based blockchains’ energy usage lack nuance, data points, and a step back to look at what these system could achieve for the world (and what price we should be willing to pay for this)

My bias

I’ve been holding various coins since 2013. I am financially incentivised for them to do well. I also believe that rebuilding the internet and financial services on blockchain-based platforms is a good thing for the world and I want it to happen. I think that because (1) storing value and having yield-generating financial services should be censorship-resistant and open to anyone with an internet connection, not just open to the richest 2/3rds of the population; (2) today’s centralised internet services are concentrating too much power into too few hands, which leads to privacy, security, and governance problems we can improve on with better architectures.

First-order effect: yes, a gigantic waste

The headlines are true:

• The very computationally expensive algorithm miners run to secure the Bitcoin & Ethereum networks (proof of work) serves no other purpose than securing the network.
• Miners get paid in Bitcoin, and the higher the price of Bitcoin, the more it makes sense for miners to invest in hardware and electricity usage, so the higher the environmental cost. (Partly offset by the block rewards halving over time)
• Because the price is open, all the miners are on the network, and all transactions are on a public ledger, it is relatively easy to estimate the energy use of the network (relative to e.g the gold market, or banking / payments institutions)
• Bitcoin is currently estimated to use 121 TWh / year of energy, about ~0.5% of the total world power, about the same as the Netherlands (lots of headlines pop up with different country names, the energy use fluctuates a lot, again because it is tied to price)
• Ethereum is believed to be around 18.9 TWh/year, which is comparable to Lebanon’s energy consumption (source)

Second-order effect: impact on energy markets

(The numbers below are from analysis and stories about Bitcoin, but I expect similar energy usage patterns from Ethereum, albeit at ~10x less volume)

However, this energy use is not all bad.

1. Some of it is renewables

• Lower-bound (that I found): Cambridge Center for Alternative Finance report that “39% of surveyed miners’ total energy consumption comes from renewables” (see page 11 onwards)
• Upper-bound (that I found): Coinshares report. ~74% powered by renewables
  • Note, this is a more biased source, they are digital asset investors, and it is older (May 2019)

2. Miners may be buying some excess renewable energy that would be going to waste, which may be helping the viability of renewables overall

• On the one hand:
  • According to both reports Hydroelectric dams is the most widely-penetrated source (not necessarily largest in terms of aggregate usage)
  • According to Coinshares, 70%+ of mining power is in China, and the biggest region is around the Sichuan hydroelectric dam where the authorities have urged miners to consume its capacity
    • “Sichuan, second only in the hashpower rankings to Xinjiang, is a province characterized by a massive overbuild of hydroelectric power in the last decade. Sichuan’s installed hydro capacity is double what its power grid can support, leading to lots of “curtailment” (or waste).” (source)
  • Interesting historical precedent: “Before Bitcoin, aluminium served this purpose. A huge fraction of aluminum’s embodied cost is the cost of electricity involved in smelting bauxite ore. Because Iceland boasts cheap and abundant energy, in particular in the form of hydro and geothermal, smelting bauxite was a natural move. The ore was shipped from Australia or China, smelted in Iceland and shipped back to places like China for construction.” (source)
  • …or stranded gas which would otherwise get burnt (admittedly anecdotal):
    • “Stranded gas is natural gas that has limited utility and is likely to be wasted. An oil or gas well that does not have the pipeline infrastructure needed to transport the gas is considered stranded. Stranded gas is flared (deliberately burned into the air to avoid the risk of explosions) or vented (allowed to escape into the air) if it cannot be used, if there is no pipeline capacity to transport it, or if prices are too low for it to be economical to transport. America’s two biggest oil fields flared and vented almost 500 billion cubic feet of gas in 2019, which would have had the climate impact of seven coal fired plants if released directly into the air. In December 2019, Crusoe Energy Systems announced it had plans to set up 70 bitcoin mining units in this year, preventing the flaring of 10 million cubic feet of gas per day. Equinor, a publicly traded petroleum multinational, also revealed plans to use stranded natural gas that would otherwise be flared and create carbon emissions to power bitcoin mining.” (source, case study)
• On the other hand
  • The Cambridge report says this buying of excess energy is likely happening with other sources than renewables. So eg coal could be being burnt unnecessarily to mine.
  • “China’s oversupply of hydroelectric energy during the rainy season has often been used as evidence in claims that a vast majority of mining is powered by environment-friendly power sources. While it is true that the Chinese government’s strategy to ensure energy self-sufficiency has led to the development of massive hydropower capacity, the same strategy has driven public investments in the construction of large-scale coal mines. Like hydroelectric power plants, these coal power plants often generate surpluses. It should not come as a surprise then that a significant share of hashers in the region equally report using both hydropower and coal energy to power their operations.“

So it’s worth keeping in mind that miners are using renewables, and in some cases miners may be buying excess capacity, so it’s not all bad.

Second-order effect: building a fairer world?

In any case, blockchains do use a lot of energy. What amount of energy use and environmental impact should we be willing to accept for their existence?

This is a bit of a “beauty is in the eye of a beholder” question. I don’t have the answer, but here are some questions I think are worth asking associated to each of benefits blockchains-based systems could bring:

• Store of value: replacing gold, but accessible to anyone with an internet connection, censorship-resistant, and easy to access, divide and transfer, and actually capped
  • What is the human cost and environmental footprint of gold production?
    • Human exploitation: gold mining was a big driver of colonisation, and there continue to be mines around the world with horrific working conditions
    • Environmental: “Dirty gold mining has ravaged landscapes, contaminated water supplies, and contributed to the destruction of vital ecosystems. Cyanide, mercury, and other toxic substances are regularly released into the environment due to dirty gold mining.” (see this report for details, it’s quite sickening)
    • Energy use: gold mining is a massive global industrial process. It’s not because we can’t easily quantify it’s footprint (until ESG reporting becomes more widespread from the mining companies) that it’s clean (or cleaner than something else)
• Building an open and more efficient financial system: enabling anyone in the world to own a non-inflationary asset, put their savings to work, borrow safely, etc
  • Might the value of building and maintaining this system (on upwards mobility and reducing inequality) outweigh its environmental footprint?
    • “1.7b people lack access to formal financial services, so are unable to securely save, borrow, or invest money, and are denied the fundamentals of financial empowerment” (source). (This is particularly bad in our world of central-bank-fuelled asset-inflation)
    • Yet two-thirds of this group are online with their phone, so they can create a digital wallet and access public blockchain-based services
    • For example, they can store Bitcoin as a store of value, have a yield-generating DAI savings account, lend and borrow using Aaeve, trade on Uniswap, etc. without anyone’s permission or approval
    • These systems can also offer better rates than the traditional financial system for users, for instance in remittences where the biggest players charge extortionary fees, or loan sharks
  • What is the environmental footprint and human cost of the current financial system?
    • … and its large workforces, offices, doing business travel etc.
    • Would it be more or less environmentally onerous overall if our financial system was run on Bitcoin, and other blockchain-based systems?
• Bitcoin as a reserve currency
  • The US dollar is currently the world’s reserve currency, which provides many advantages to the US, e.g “The dollar’s centrality to the system of global payments also increases the power of U.S. financial sanctions. Almost all trade done in U.S. dollars, even trade among other countries, can be subject to U.S. sanctions, because they are handled by so-called correspondent banks with accounts at the Federal Reserve. By cutting off the ability to transact in dollars, the United States can make it difficult for those it blacklists to do business.” (source)
  • What is the human cost and environmental footprint of maintaining the dollar as the world’s reserve currency?
    • OK, this is getting too speculative and way above my pay grade, but you could make the case that much US imperialism in the 20th century was indirectly driven by maintaining the dollar’s hegemony…

Proof of Stake

You likely know about Proof of Stake, the alternate consensus algorithm which is significantly less wasteful than Proof of Work (I’ve not seen any estimate but would say at least 1000x less wasteful given mining network sizes)

Proof of stake is live and working in production in the Tezos, Solana, Cosmos and Neo blockchains to name a few. It’s on a test version of Ethereum and expected to be rolled out to the Ethereum “main-net” in the next year or two.

It’s possible that Bitcoin will eventually follow and also move to it, I don’t know how likely that is.

If it doesn’t, it’s also possible (and I think would be great) that Bitcoin gets displaced in favour of another blockchain which can also provide the same benefits, but with a proof-of-stake consensus algorithm. This could be accelerated by e.g a carbon tax in the countries which have a lot of miners.

Conclusion

Bitcoin’s energy use can easily be measured, so it’s easy to point the finger at.

We should be comparing it to the environmental (and human, and financial inclusion) costs of the systems it might displace, which are harder to measure.