Several reports and videos out of China over Thanksgiving weekend have shown large-scale shutdowns of bitcoin mining equipment, as the falling price of bitcoin made mining unprofitable. Broader data show that miners worldwide are going offline. This might seem incredibly scary when viewed through a traditional business lens: When Sears shuts down yet another batch of stores, for instance, it’s a pretty clear sign that Sears as an entity, business, and even as a concept is headed for the dustbin of history.
But that’s not how bitcoin works, thanks to one of the many strokes of genius baked into its fundamental design: difficulty adjustment. The bottom-line explanation of difficulty adjustment is that every time a mining rig is shut down, the bitcoin protocol increases the incentive for other miners to stay online. It’s as if every time a Sears shut down, all the remaining stores became more profitable.
Gazed upon long enough, difficulty adjustment can take on a kind of transcendental religious quality. It’s not the reason bitcoin is useful day-to-day, but it is the reason that bitcoin (like other true blockchains, and very much unlike Sears) is nearly immune to true destruction, whether by state regulation or market fluctuations. It’s the reason bitcoin is superior to a centralized digital currency system like eGold, even if bitcoin is more expensive to run in aggregate. It’s what makes blockchains more persistently “real” than mere data saved to a hard drive, or even, arguably, to a Google cloud server farm.
To understand difficulty adjustment more deeply, let’s start with a quick refresher on how mining works. In order to add a block of transactions to the bitcoin ledger, a mining node has to find the solution to a cryptographic problem. The node that successfully finds the solution is rewarded with new bitcoin, currently 12.5 coins, or a bit under $50,000.
This process, more formally known as Proof of Work, is also sometimes referred to as “solving a cryptographic puzzle.” Calling this a ‘puzzle’ or ‘problem,’ though, is fairly misleading; solving a “puzzle” often relies on some sort of logical reasoning, but the solution to a bitcoin block is essentially randomly generated by the bitcoin protocol. Mining rigs are making very complicated, equation-backed guesses as fast as possible, in the hope that they’ll be the first to hit the random solution and reap that sweet, sweet, reward.
This randomness is key to difficulty adjustment. Every two weeks (or, more precisely, every 2016 blocks), the bitcoin protocol adjusts the difficulty of finding blocks by changing the range of the possible (random) solutions that miners are trying to guess. To increase difficulty, the protocol makes the range of possible solutions wider, increasing the time it takes for all those miners to collectively guess the right answer. To decrease difficulty, the range is made smaller. Think of difficulty adjustment as changing the size of a target that a bunch of extremely high-tech (but ultimately quite dumb) monkeys are trying to hit with a rock.
The code adjusts difficulty according to the amount of mining power on the network. When there’s less mining power on the network and competition to solve blocks declines, difficulty is reduced. When there’s more power and competition, it’s increased. The protocol’s goal is to find a difficulty point at which the network solves a block and settles outstanding transactions every ten minutes. But the effects for miners are more concrete; by decreasing difficulty, the bitcoin protocol also decreases the amount of time, processing power, and electricity required to solve a block.
That’s exactly what has been happening since mid-October, when difficulty dropped about five percent, according to Bitcoin Wisdom. Then just last week, it dropped again, by nearly 7.5 percent. Bitcoin Wisdom currently projects the next reset will be a whopping nine percent drop.
If you’re lucky, understanding difficulty adjustment is one of the moments when the struggle to truly grasp bitcoin yields that delicious galaxy-brain whoosh of witnessing sublime brilliance. Because when a big wave of miners shuts down or goes offline, as they now seem to be in China and elsewhere, declining difficulty in turn increases the relative reward for miners who stick around. A difficulty decrease, on average, reduces the cost to mine a block, and increases miners’ ability to make a profit at a particular price, with a particular piece of equipment.
How this impacts the profitability of bitcoin miners or mining operations is complex, since it doesn’t change costs such as rent, maintenance, and payroll. But it does directly decrease the electricity cost of mining, recently estimated at $2,500 per bitcoin. And while older equipment may be shut down in response to price declines, a downward difficulty adjustment could make that equipment relevant again—or at least slow down the arms race towards ever more expensive, high-powered equipment.
I’m leaving out a lot of unknowns and complexities about the impact of price and difficulty declines, some of which are poorly understood because difficulty drops have been pretty rare over the last ten years. Even in January and February of this year, as bitcoin prices were plummeting, difficulty was going up because mining was still very profitable, continuing to attract more computing power. So, in some sense, the downward difficulty adjustment is a strong signal that we’ve entered the true depths of Crypto Winter.
But at the same time, it’s a reminder that bitcoin has a built-in safety net. Even as the market price of a bitcoin craters, the bitcoin network is getting cheaper to maintain. That might or might not help bitcoin survive catastrophic climate change or a Mad Max-style social collapse—but its elegant socioeconomic engineering, if nothing else, is eternal.