Bitcoin’s Energy (ab)use: Smells like Dung. But Horse Dung

Ah! the 1800s!

The melody and quaint sight of hansom cabs, horse-drawn buses and goods-ferrying drays.

Horses did change a lot back then – lanes, traffic signals, culture, social strata and clothes.

Until people in London and New York overlooked one crucial upshot of so many horses (a horse produced between 15 and 35 pounds of manure per day on an average) – Horse-crap and miles of it.

After slowly, but strongly, turning dependent on horses for transport needs, many cities by the late 1800s found themselves stuck beneath piles of horse manure. Yes, the Great Horse Manure Crisis of 1894 taught us some totally-unexpected-unwieldy-and-disproportionate side-effects of something as useful, as innocuous-looking and as convenient as horses.

The lessons did not just incommode noses but human lives too. The manure cascaded into far-reaching and far-worse ripples – Typhoid fever. Decaying horse carcasses. And predictions that were hard to shrug – “In 50 years, every street in London will be buried under nine feet of manure.”

Thankfully, ways and wheels of new kind were found and the epidemic averted.

Today, when we hear the clippety-clop of a new creature in our financial streets, it seems as harmless and as exciting as a horse-carriage; but should we start exerting our noses already?

Picture these fears:

Bitcoin mining worldwide uses more electricity than all of Denmark or Ireland. Soon enough, bitcoin could be consuming some 1.8 per cent of global electricity (Defeating the progress made by renewable sources so far). Researchers like Alex de Vries, from the Experience Center of PwC in the Netherlands – reckon the minimum current usage of the Bitcoin network at 2.55 gigawatts. In other words, a single bitcoin transaction is so energy-intensive that it could power the average U.S. household – for a month! Or to extend it further – the network could be using as much as 7.7 gigawatts–as much as Austria. Or maybe even half of a percent of the world’s total consumption.

Bitcoin’s energy consumption can be as much as 23 terawatt-hour per year (as per January 3, 2018 Morgan Stanley) (Image courtesy Danilo Rizzuti Freedigitalphotos)

Yes, bit coins, Blockchain and Crypto-mining. Not strangers to be introduced anymore.

They have created as much noise on market roller-coasters as they have done under the hood of electricity-grids. They have a lot to offer, for sure, in terms of their quintessential advantages of anonymity, transparency and immutability; but there is so much that needs to be figured out when it comes to what they poop.

Every recent sign around – from Google Chrome extensions being fiddled for crypto-jacking computers for mining Monero; to huge botnet attacks seen in India, Russia and Taiwan (half a million machines were infected with a crypto-mining botnet); to 2018 estimates from HS Orka of how Icelandic data centers mining crypto-currencies are gobbling 840 gigawatt-hours (GWh) of electricity; to International Energy Agency’s worries that the entire Bitcoin network consumes more energy than a number of countries.

Wait, why this much power? It’s mining right? So how can it be worse than gold mining?

For one, like gold mines, bitcoins too exist in the protocol’s design (underground and may be 21 million bitcoins will exist at some point), and miners work on computers to bring them out into the light, a few at a time. This happens via guessing a random number and this hash function work (miners get rewarded with crypto currency for solving a Maths puzzle basically) – ensures the innate strength of security and decentralisation of blockchains, but alas – can wreak havoc on computing power resources (imagine 10 minutes of computer time to process one block). Time-stamping helps induce credibility in the system so that no frauds can happen, but that translates into what they call as nonce and hence, energy-hungry miners.

Of course, advocates in favour of this technology say that most power-consumption figures are inflated. They also point out how the number of bitcoins awarded as a reward for solving the puzzle will move downwards (12.5 now, but halving every four years with the next one expected in 2020-21). Another contention is that the value of bitcoin relative to cost of electricity and hardware will also play a role in compensating this reduction.

All this also needs to factor in miner’s incomes and their correlation to electricity costs, the chance of underestimating the network’s energy consumption, and ignorance of factors like machine-reliability, climate and cooling costs.

What if we could work around Proof-of-work energy issues? Or use idle mining time better?

There have been some objective peeks though – like how the initial method of Proof-of-Work (to take out the possibility of any dishonesty and fraud in a blockchain scenario) needs to be fixed to bring down the current heavy-compute-usage. BOINC, Primecoin, Gridcoin, MangoCoinz and SolarCoin were some more projects rolled into this world to make more use of the computer-power and idle time that miners spend here. Same is the idea behind virtual mining.

But how much can hopes and raw attempts help? Is there a Henry Ford coming to rescue us from horse-crap this time? We had to ask someone. So why not someone who has the guts and the grey matter to say this:

“..someone who goes around cutting people’s hands off should be praised for helping the field of prosthetics.”

Yes, the same voice that sardonically attacks bitcoin-good-news guys. “… there is no thing that is not good news for Bitcoin. Exchanges kicked out of China? Good news. Price crashes? Good news…did you know that Bitcoin using 0.1 per cent of all the electricity on earth, and 0.5 per cent by the end of the year is good?”

David Gerard, author of “Attack of the 50 Foot Blockchain: Bitcoin, Blockchain, Ethereum & Smart Contracts”, helps us understand this conundrum (all its what-ifs, may-be-s and hell-what-s) with a nose that is smelling crap amidst all the pom-poms. He answers why we cannot soothe ourselves with sprays like ‘oh, it is just a bubble’; ‘well, wait let compute-parts get better and more efficient and then, watch’; and ‘the pleasant irony is that bitcoin will actually spur more eco-friendly energy use’.

Miners will keep going, bubble or bust: David G

Let’s take off those blinkers, at least, for a few minutes here, then.

Should we be even worried about bitcoins and the energy-guzzled here? What if it is another bubble that will fizzle out soon? How serious and out-of-control is the energy-usage problem owing to virtual currency, blockchain and such new-age technologies?

The problem with crypto-currency mining is that it’s designed to use more and more energy, even while only doing its work – the transactions – at the same rate. So it will always use more and more. We see this already in Bitcoin.

When the 2013 Bitcoin bubble popped, the miners kept on going. When prices dropped through 2014 and 2015, the miners slowed down the rate of growth, but never went backward. So as the present bubble fizzled, there’s no reason to think they’ll stop this time either.

Is it about the inherent Mining process or the Consensus issue or the way it is chosen to be done? Any alternatives?

Proof-of-work is the big problem. The “work” is literally wasting energy; you prove how much energy you can waste to demonstrate that you should win the bitcoins.

Another consensus mechanism that doesn’t just waste power might help. But there are theoretical reasons to think that you can’t do a fully decentralised crypto-currency unless you are wasting power, because to attack the chain successfully requires that many resources – so their horrible wastefulness is their security. So the real solution is to stop using crypto-currencies.

Is the Proof-of-work vs. Proof-of-stake debate apt here? Also, what fares better – Public or private or hybrid blockchains? Would sharding and side-chains help a lot to address energy-angles of blockchains?

Private blockchains don’t use proof-of-work – since they’re private, with known participants, they can use various methods of just-taking-turns.

Proof-of-stake may help, particularly if we can get a crypto-currency that gives a platform that participants are willing to use.

I think proof-of-stake blockchains will naturally centralise – but this gives a more stable platform for the participants. Bitcoin maximalists are very upset by the idea of this – but the crypto trading market treats proof-of-work currencies like bitcoin, completely centralised currencies like Ripple, and centrally-administered ICO tokens as the same sort of thing – “crypto”. So the market does not care about bitcoin ideology, only market ideology.

0.1 per cent of all the electricity in the world for a tiny number of transactions per second – a horribly inefficient currency: David G

What about experts and companies who say that evolving computing power and efficiency-innovations would solve the problem soon? What’s the relevance of FPGAs, ASICs and graphic cards here? Does use of hydroelectricity or spare electricity (China) or renewable energy-grids make the implications less worrisome?

Experts who say this do not understand how proof-of-work works – if you make the parts more efficient, you’re still putting efficient parts in inherently wasteful competition.

Bitcoin mining progressed through CPU (Central Processing Unit) mining, video-card mining, FPGA  (Field Programmable Gate Arrays) mining and the current ASIC (Application-Specific Integrated Circuit) mining – each more efficient than the previous one. But the efficiency gains were only for miners to gain a temporary advantage over other miners – they did not make the whole system any less wasteful.

Renewable energy may help – but crypto mining requires 100 per cent 24 hour a day operation to be economically viable. This is a very bad load for wind or solar. And hydroelectric power is limited, and is frequently already apportioned to towns.

Cheap Chinese hydroelectric ones used to be badly connected to the national electrical grid – so the power was just being wasted. This is increasingly not the case, however – and as hydro plants are hooked to the national grid, the bitcoin miners are being told to stop using them.

Why do these technologies lack economies-of-scale at a fundamental level? How does this compare with other currencies/gold-energy footprint?

The key difference is that in every other technology, all the incentives are to become more efficient – since your production costs are a cost item, and you want to reduce that cost.

With Bitcoin, it is ‘anti’-efficient – all the incentives are to use more and more power, for a system that keeps processing at only its original fixed maximum rate.

What’s your take on the idea that – More bitcoin-usage incentivizes greener planet efforts in the big picture, eventually?

This is false. Bit coin mining requires 100 per cent 24 hour a day usage – this is a load optimised for coal, not for wind or solar. It’s just a bad statement Bitcoin advocates make to try not to sound as bad.

This is like saying that someone who goes around cutting people’s hands off should be praised for helping the field of prosthetics.

How does Crypto-jacking fit in here? Would the rise of IoT make things better or worse?

It doesn’t fit so much into this, except that PC users are learning to notice immediately when they’re crypto-jacked, because their PC fans suddenly turn on full and the PC gets hot.

Does the energy-question contradict the scale and security aspect of mining in most cases?

The energy wastage is literally what proof-of-work uses for its security – so that you would need to spend that much energy to attack it. The actual solution is that proof-of-work crypto-currencies are vastly wasteful, and crypto mining should be severely restricted. Spending 0.1 per cent of all the electricity in the world for a tiny number of transactions per second in a horribly inefficient currency is a bad enough externality that it’s becoming everyone else’s business.

Pratima H


Crypto Canon


Leave a Reply

%d bloggers like this: