Why Right Now is one of the Best Times to be Running a Bitcoin Miner
Nick Slaney is a long time bitcoin user, node runner and advocate. He currently works on hardware at Square as a Product Design Engineer. The views stated below are his own and not those of his employer.
While it is very common in 2021 to speculate on what most recently caused the price of Bitcoin to drop (even though the price is still up significantly from 2020), I see less people talking about what is increasingly becoming a hot topic if you have any inclination towards mining: it’s a crazy time to be running a bitcoin miner right now (as long as you’re not in China).
Miners (note: in this article I will call people who participate in bitcoin mining “miners” and the machines that do the work “mining hardware”) keep the bitcoin network going by competing to add blocks of transactions to the blockchain. To earn the right to make the next block and claim that 6.25 BTC reward (+ transaction fees) miners use specialized hardware devices that frantically try to find the correct solution to the next block based on the previous one. A mining rig’s typical output is measured in hashes / second and current ASIC based mining hardware can crank out anywhere from 0.1–110 terahashes per second per rig.
There are many considerations to becoming a bitcoin miner including finding and buying mining hardware, that hardware’s efficiency, a space to hold that hardware, and, as we see current events unfold in front of us, the political environment in the region where you do business. All of that being said, the largest cost and consideration by far for miners today is the cost of energy. The cost of the energy to run your mining hardware determines whether you make money, break even, or lose money for the sake of creating your own sats.
Knowing a bit about the way mining works, we can actually put together an equation that looks at mining profitability depending on the amount of hashrate you have, the efficiency of your hardware, total hashrate of the network, and your energy costs. Here’s a peak at that equation looking at potential profit in USD per day:
To show you how crazy it is to be a miner right now (and how it has been for most of 2020), we can rearrange this equation to find the break even cost for the entire network to create a bitcoin.
To be clear, we have to make some assumptions here to approximate the current energy cost of a bitcoin. Namely excluding the factors that go into a successful mining business that I mentioned before. Also be cognizant that these factors can change over time, especially the price of bitcoin and total hashrate! That being said, I think it’s fair to say the constraints I’ve picked are on the conservative side, and we can plug in some real numbers from public mining companies to see real world cases as well.
I’m also leaving out the profit aspect of this equation for a simple reason. In the past it’s been assumed that miners will sell mined bitcoin to cover energy costs, but that isn’t always the case. If you’re a miner and have the capital and desire for bitcoin, running your hardware at break-even and even for a loss may not be out of the question. You may be willing to pay a premium for coins that don’t come with KYC documentation attached like what you get on Coinbase, or you may be very bullish on bitcoin and OK with taking some loss if you think the price will go up substantially from here.
Regardless of your view of what constitutes an “acceptable” amount of profit for the capital you risk in hardware, facilities, etc. I think what you’ll see here will convince you that profit, at least at this current moment in time, is not very far out of reach.
The Energy Cost of Bitcoin
So let’s rearrange the equation to solve for the total network energy cost of one bitcoin and justify the assumptions for the numbers we’re using. First the equation:
Note this is based on a single day of bitcoin mining, where given the current block reward of 6.25 BTC, on average we see about 900 BTC created. (If you’re curious about how I derive these equations, let me know. I’ve been working on a longer article breaking them down, but this one seemed more timely to get out in the meantime.)
The assumptions:
- For total hashrate (TH/s) we’ll use the current average rate today of 87 exahashes / s or 87,000,000 TH/s
- For estimated miner efficiency, I’m pulling information from two North American mining companies, Riot and Hut8. 32.3 and ~100 J/TH (HUT8’s figure is imputed from their power capacity and current TH/s figure, thus it may possibly be an overestimate) respectively. Let’s split the difference at 70 J/TH. Miners can have a range of hardware, from old to new with a range of efficiency. Note the most efficient known hardware available now is the Antminer S19 Pro at 29.5J/TH.
- For energy cost, I’m using a “retail” rate that is typically seen if you co-locate your hardware with bigger miners in the US, $0.06/kWh. This rate is a decent cost of energy that also implicitly includes maintenance and facility costs as well as the comfort of a politically accepting region. Miners like HUT8 and RIOT likely get much lower energy costs on the order of $0.02–0.04 / kWh, but invest in large facilities and energy contracts on the order of megawatts to get them.
- We are not amortizing the cost of mining hardware here. Hardware costs fluctuate with the price of bitcoin substantially.
Phew, so this isn’t the perfect model of what a bitcoin costs, but I think it’s close. It’s actually likely that the cost for most large miners is lower than the numbers we’re using.
So what’s the (conservative) cost of a bitcoin in energy terms right now?
$9,744
Considering they’re selling at ~$34k right now, it’s not a bad deal if you ask me.
I did the math on Riot’s current daily profit in USD as well. 16.55 BTC or $500k. That’s per day.
How did this deal get so good?
Typically when the price of bitcoin goes up, the profitability of mining goes up as well. All things in our first equation held equal, if you’re making more on those coins, your profit is going up.
But, any miner worth their salt is constantly looking at the price of bitcoin and their own costs, and when they see the price coming up they may bring previously unprofitable hardware back on line, as well as get investment to bring up more mining hardware. Typically, this leads to a gain in total hashrate, which through biweekly difficulty adjustments dilutes all miners’ share of the static 900 BTC generated per day, raising the cost of bitcoin and to an extent regulating the profitability of miners already on the network.
In previous bull runs, this effect of hashrate following price can be seen. As hype builds and there is money to be made, and as it falls, miners with older, less efficient hardware and/or higher energy costs see their profits dwindle and turn their miners off.
The 2020–2021 bull run has been different though. Lockdowns put in place to slow the spread of COVID-19 have wreaked havoc on businesses and supply chains in particular. What would’ve been a year of huge production for mining hardware makers has been plagued by lockdowns slowing production and shortages of a major component in mining ASIC production: silicon, the effects of which are projected to be felt until 2023 at least.
As a result of all of this, as the price has risen, miners have been desperate to get their hands on hardware, but haven’t been able to get it. The market for second hand hardware certainly shows it, the market price of an S19 Pro that was ~$2k in 2019 now being $10k+. Hardware producers like bitmain are currently only selling hardware to customers with multi-million dollar contracts and even then, those customers are only getting a slow trickle of hardware.
To add to this already unprecedented situation, the Chinese Communist Party, who have been known to be a bit anti-bitcoin in the past, now seem to be serious on cracking down on bitcoin mining in China. It’s hard to tell the exact motives or situation in China without being there and speaking mandarin, but the effect it has had since the announcement in April has been dramatic.
Hashrate has fallen nearly 50% at a time when our model points to the profitability of mining being absolutely huge. The last time the hashrate was this low, a bitcoin was actually $10k, and mining was less money printing and more a venture for those who believed in the fundamentals and future price potential of BTC. Well that future is now, and for anyone who’s lucky enough to be running mining hardware, their initial investment is paying off substantially.
Will it last? Personally I think the hit to hashrate from the CCP’s actions has been dramatic and is likely temporary as mining hardware moves out of the country, but silicon supply shortages are real and will be a factor for the next couple of years. If we look back to April right before the crackdown in China, even at a network hashrate of 160 EH/s our conservative estimate still puts the cost of a bitcoin at $20k, with a nice amount of profit to be made, especially at the recent highs of $64k. All I know is, anyone who is running mining hardware and not living in China is probably smiling a bit more as of late.
