Recently, I was invited to give a talk at the Texas Blockchain Summit on the topic of the growth of bitcoin mining in Texas. Not knowing anything about bitcoin mining in Texas, I interrogated around two dozen mining entrepreneurs, wholesale energy traders, academics and energy experts. What I discovered would completely alter my views on bitcoin mining.
Put shortly, bitcoin mining is converging with the energy sector with amazing rapidity, yielding an explosion of innovation that will both decarbonize bitcoin in the medium term, and will dramatically benefit increasingly renewable grids. What’s more, it appears that only bitcoin – rather than other industrial load sources – can actually achieve some of these goals.
Bitcoin mining is suitable for renewables – and the proof is in the pudding
Bitcoin enthusiasts have long maintained that bitcoin mining could drive a clean energy transition. The logic went as follows: Bitcoin miners voraciously buy the cheapest sources of energy available; renewable (wind and solar) sources of energy are getting progressively cheaper and will soon outmatch thermal energy on cost; bitcoin miners will therefore subsidize the buildout of renewable energy, benefiting everyone.
I was long suspicious of this argument, due to the acknowledged low capacity factors exhibited by wind and solar sources of energy. Given that miners enter a race against time to get the most out of their new mining machines before hashrate climbs and they become uneconomical to run, putting those miners to work with low-uptime solar or wind energy wouldn’t make sense. Indeed, this was a common critique leveled at bitcoiners promoting this narrative: that renewables weren’t suitable for bitcoin mining.
However, a few new developments within the mining market completely changed my mind on this. I now firmly believe that bitcoin mining is suitable for and beneficial to renewable energy, both on a first order and second order (indirect) basis. The key developments that changed my view are the following:
The emergence of the “lifecycle mining” concept
The development of a new hybrid model for bitcoin mining that is partially grid-based and partially “behind the meter” mining
The slowing of ASIC development cycles.
Lifecycle mining is a concept I learned of at a talk given by Ro Shirole of miner Compute North at the Bitmain Mining Disrupt summit in Miami. In short, it refers to the concept that the energy inputs and datacenter model that miners employ should be tailored to the age of the hardware. Effectively, the practice of mining is becoming heterogeneous across hardware vintages: the type of energy deployed depends on how old your machines are. Newer units generally go in higher-assurance data centers, and get plugged into reliable energy with high uptime assurances. This is generally grid energy and inherits its carbon intensity – normally a mix of high and low carbon sources. If you have latest-gen units, you want to take advantage of them right away before hashrate climbs. Thus you can afford to pay a bit more for power, because they’re more profitable.
For older units, like Bitmain S9s, which are five years old at this point and still represent 30% of the network, according to Coin Metrics, the considerations are quite different. These units head to “ASIC retirement homes,” where they can take more interruption. Operators with these units look instead for dirt cheap power. If the economics make sense, they will run them.
Read More: The Frustrating, Maddening, All-Consuming Bitcoin Energy Debate | Nic Carter
Thus while grid energy – or more stable, low-carbon sources like nuclear or hydro – is most suitable for high-end ASICs, more intermittent renewables, even at a capacity factor of 70%, make sense for older units. The opportunity cost is much lower with older units, so they can be placed alongside a more intermittent renewable source of energy and still be economical.
Additionally, a new hybrid model for Bitcoin mining is emerging. Bitcoin miners can buy energy from energy providers when energy is abundant (as in West Texas with an excess of wind and solar), drawing from the grid the rest of the time. In so doing, the miners monetize a renewable asset that would otherwise be dumped into the ground, while maintaining generally high uptime. During periods of energy scarcity, the miners can be turned off. The net effect is that renewables become more economical, as they can monetize their asset even when the grid has no demand for it.
Critics either don’t understand this concept or deliberately misrepresent reality. For instance, Popular Science claims that “there are no excess renewables,” citing an academic stating “if bitcoin miners or cryptocurrency miners are going to take that renewable, that means it’s not there for somebody else to use.” This is blatantly untrue, as demonstrated by hydro power use in Sichuan. In Texas, the western portion of the grid boasts 32 GW of capacity (much of it wind and solar), 5 GW of load and only 12 GW of transmission to load centers elsewhere in Texas. The rest of the power is habitually curtailed, and it’s no wonder that bitcoin miners are flocking to the region.
Alex De Vries, the notorious Bitcoin energy critic (who created Digiconomist and works for the Dutch central bank) claims that miners are “the ideal customer for obsolete fossil fuels rather than renewables since these are both cheap and a source of constant power.” One wonders how he will react to news of low-carbon energy providers engaging with bitcoin miners to serve as a source of additional offtake.
Already, we know that hydro (which gets curtailed seasonally) has historically been a huge source of supply for bitcoin miners, and remains one, in Canada, Russia, Washington state and New York. Now, miners are signing up deals with nuclear power plants (which often produce excess power at night, when the grid is less demanding). It’s only a matter of time before solar and wind energy producers begin to take advantage of bitcoin mining as an additional, uncorrelated buyer. Already, we see Norwegian energy giant Aker declaring their intention to use bitcoin as a “load-balancing economic battery” and to become a “valuable partner to new renewable projects.” Blockstream and Square are partnering on a solar-powered mining facility.
Lastly, as ASIC releases get more infrequent – and they are undeniably slowing down, due to the physical limits being reached at the transistor level – miners are more incentivized to find cheap energy, rather than racing to get the latest units active. Despite the absolutely comical claims by critics that ASICs last only 1.29 years, ASICs are lasting longer and longer, and new cycles are less and less frequent. This means that miners can take a longer-term focus and work to find cheap, renewable sources of energy.
Flared gas mitigation is as close to a free lunch as you can get
I first heard about mining as an alternative use for waste methane from Steve Barbour of Upstream Data, who is acknowledged as having pioneered the concept in 2017. For a long time, it was considered a bit of a pipe dream, destined to remain a highly niche sector in the mining space. However, today there are a number of well-funded companies actively deploying assets in the flared gas mitigation space. These include Giga Energy, Crusoe Energy, Great American Mining, Nakamotor Partners, Jai Energy and Upstream.
Mining with waste methane at oil wells makes perfect sense because it is a natural byproduct of oil extraction, especially during “initial production” when you get a huge, short term burst of gas. Many oil wells are completely remote from pipeline infrastructure, and due to the prices of natural gas, do not find it economical to transport the methane to refineries. Thus they simply opt to flare the gas (venting it would be far worse, as raw methane is a much worse greenhouse gas than CO2, the output of the combustion). But flaring is inefficient, and on windy days, much of the methane is not consumed in the reaction.
What bitcoin miners do instead is capture the natural gas, pipe it into a generator present on the well pad, and use that energy to power bitcoin miners. This is a more complete, controlled burn and therefore reduces emissions associated with the alternative in which the gas is flared. The systems deployed in the field are highly modular and transportable – if a well is producing an abundance of natural gas following initial production, miners can come in for the first six months and take advantage of that initial burst, which wouldn’t otherwise be captured (since no operator would build a pipeline for a short term glut of cheap gas).
It’s no surprise that Texas Sen. Ted Cruz extolled the practice in his recent comments at the Texas Blockchain Summit:
“Fifty percent of the natural gas in this country that is flared, is being flared in the Permian right now in West Texas. I think that is an enormous opportunity for bitcoin, because that’s right now energy that is just being wasted. It’s being wasted because there is no transmission equipment to get that natural gas where it could be used the way natural gas would ordinarily be employed; it’s just being burned.”
The common response to the flared gas mitigation use case is that we shouldn’t be extracting any hydrocarbons, period, and thus waste gas miners are still producing illegitimate emissions. This talking point is aimed to delegitimize the excellent work that bitcoin miners working with otherwise flared gas are doing to manage this waste product in an efficient and low-emission way.
But this is a fundamentally Malthusian stance: We are nowhere near a move to a non-fossil fuel energy standard, and to do so would be suicide on a civilizational scale. Cut off natural gas, for instance, and billions of individuals will not be able to heat their homes in winter. Curtail petroleum use, and our transport system will fail. The globalized system of trade would collapse. Farmers, lacking fertilizers, will not be able to feed the earth’s population. Death and famine at a mass scale would directly follow an abrupt end of oil and gas extraction.
For now, until we invent nuclear fusion, alternative fuel types or some other energy golden bullet, oil extraction will remain a civilizational necessity. If the critics aren’t willing to live in the counterfactual world where we cease using hydrocarbons, the seriousness of the critique should be questioned.
The other objection to flaring is that it should be regulated out of existence, as many states in the U.S. have done. But this wouldn’t stop flaring – all it would do would be to make hydrocarbon extraction more expensive and unwieldy in the U.S., advantaging oil extraction abroad. A ban on flaring, if it raised costs for producers, would simply cause the U.S. to import more oil from overseas.
Foreign states like Saudi Arabia and Nigeria have no qualms whatsoever about flaring. As long as the world has a need for oil – and this won’t change anytime soon – there will be waste gas produced at the well pad. Bitcoin miners are undeniably the best positioned to sustainably mitigate this waste product – and also protect a global monetary network while they’re at it.
Flexible load allows grids to accommodate more renewables
Lastly, and perhaps most importantly, a new feature of bitcoin mining has begun to gain traction, to the benefit of power grid operators. Bitcoin miners represent “interruptible load,” which means that they can deal with power outages without suffering adverse impacts to their business. Of course, they prefer to have power all of the time, but nothing catastrophic happens when they lose power, unlike other industrial consumers such as hospitals, high-end data centers, factories and smelters.
This makes them perfect for so-called “demand response” programs, which refers to formal or informal agreements to curtail their demand when the grid is overtaxed and prices are high. This means that when energy is in short supply, bitcoin miners can turn themselves off and get power to the households that need it most.
In exchange for agreeing to have their power interrupted some percentage of the time, miners get rebates, so it’s economically worthwhile to opt into these programs. Even if miners aren’t participating in an explicit rebate program, surging market prices during shortages in deregulated grids like ERCOT send miners the signal to turn themselves off.
During his remarks in Austin, Texas, Sen. Ted Cruz demonstrated his familiarity with the concept, pointing out the benefit of having bitcoin miners on the grid as a source of interruptible load.
Even more compellingly, miners can serve as a source of “controllable load,” which is a concept I learned from Lancium. This means that miners can can dial up and down their consumption to a level the grid operator demands – within seconds. Instead of tweaking the supply side, like turning on gas turbines to make up for a sudden interruption of wind, grid operators can instead ask mining data center operators to dial down their consumption. Having this option means fossil-fuel powered peaker plants don’t need to be triggered as often.
The flexibility of these miners is not going unnoticed. The 2020 state of the market report from the ERCOT independent market monitor marvels at 100MW worth of new data centers (all mining Bitcoin) using “fast acting control systems” that enrolled as Controllable Load Resources. Grid operators have rarely encountered such flexible consumers of energy, and it has taken them some time to design appropriate programs to take advantage of these resources.
More demand-side control for grid operators means fewer carbon-intense peaker plants. And due to the growing influence of renewables on grids like that of Texas, more flexible load is welcome. Wind and solar, unlike coal plants, hydro or nuclear, don’t produce energy reliably. Their intermittency means they need to be backstopped by batteries (mostly uneconomical at present), or gas-powered turbines.
More controllable load however alleviates this intermittency without requiring more fossil fuels. A new paper from Texas energy academics finds that more controllable load would actually assist in decarbonizing the grid. In their words, “Operating the data centers in a flexible manner during times of high grid prices could lead to the deployment of even more wind and solar and – if they are operated with enough flexibility – could result in lower overall carbon emissions.”
During the conference, Cruz compared bitcoin mining to fracking, another innovation that has been much maligned by environmental activists, but has contributed to energy independence in the U.S. and actually led the decarbonization of the U.S. grid, as natural gas has around half the carbon intensity of coal.
His comments were apt: Bitcoin mining appears to be wholly consistent with the goals of environmentalists in the U.S., as it safeguards grids made unstable by new wind and solar assets; monetizes hydro and nuclear when the grid is not a buyer; and settles into off-grid niches like waste natural gas. That he has achieved this level of sophistication on the topic is truly remarkable, as bitcoin is not a policy priority of his. One can only hope that the rest of his Senate colleagues do the same.