Insights

Why Bitcoin Miners Will Keep Mining

By   Christopher Bendiksen 17th March 2020
  • Bitcoin prices have dropped sharply, cutting mining revenues in half even before the actual incoming halving
  • This situation will undoubtedly bring the ‘mining death spiral’ argument back to the forefront of the bitcoin media cycle
  • ‘Mining death spirals’ do not actually happen in real life––they are highly theoretical edge cases without any historical real-world precedent
  • Mining cost is largely a function of the difficulty, this is a dynamic metric determined by the protocol itself and it can adjust both up and down to keep block times at 10 minutes on average
  • This will be challenging for high-cost miners and many will not survive
  • If prices do not recover the hashrate will fall––and when the halving hits it will fall again––this is not a problem for Bitcoin, nor is it unprecedented


Glossary

  • Bitcoin - The Bitcoin protocol, network and monetary system
  • bitcoin - The bitcoin monetary unit
  • Hashrate - The total current number of hashes performed per second by the mining network
  • Mining Difficulty - An automatically adjustable limit which regulates the cost of mining


For the Miners, This Price Move is Effectively Equivalent to the Halving

Since its February 2020 peak of around $10,500, the bitcoin price, and with it, the mining reward, has fallen by more than half. And in a way, for all of those who are worried about the halving this is a perfect prelude because the end effect on miners is the exact same. Hence, the hashrate dynamics we’re likely to see in the upcoming weeks will be an excellent parallel to those we might also see after the halving in May.

So what are we likely to see? Reductions in hashrate. And that’s totally fine.

Here’s the how and why:


The Mining Difficulty Adjustment Keeps Bitcoin Block Frequency Steady, No Matter the Amount of Total Network Hashrate

Miners rely on the mining reward to cover their ongoing electricity costs. The mining reward is made up of two parts: 1) Transaction fees; and 2) Newly created bitcoins. Miners have costs denominated in their local currency, so the purchasing power of their bitcoin income is dependent on the exchange rate between bitcoin and their local currency.

When the bitcoin price falls, the miners’ bitcoin income offers less purchasing power to cover ongoing electricity costs. As a result, the miners with the highest production cost will no longer be profitable and will be forced to stop mining (more on these dynamics here, reading it will make this article a lot easier to fully grasp). This is almost exactly analogous to normal commodity production cost curves.

The overall network hashrate will then drop, but this is where the comparisons to normal commodity production stops. For example, iron production costs are the same no matter the price of iron, whereas the production cost of bitcoin is variable and dynamic (see accompanying graphic for an idealised illustration).

Every 2016 blocks (approximately every 14 days), the Bitcoin protocol adjusts the difficulty to reflect the average hashrate of that period. The rule is simple: Bitcoin blocks should take exactly 10 minutes on average to create. If the 2016 blocks came every 9 minutes on average instead of every 10 minutes, the mining difficulty is increased. If blocks came every 11th minute on average the difficulty is reduced.

Therefore: No matter how little or how much hashrate the combined mining network produces, the Bitcoin protocol will automatically adjust the difficulty such that new blocks are found every 10 minutes on average.



Source: CoinShares Research

Remembering that new bitcoins are created every block, the difficulty adjustment ensures that no amount of added hashrate could make bitcoins be produced any faster than prescribed. The opposite is also true.

An important net effect of this, is that the cost of mining always tends towards the market price of bitcoin. If the price of bitcoin increases, more hashpower will be brought online. This raises the difficulty until the cost of mining again approaches the price. If the price of bitcoin decreases, some hashpower will be removed from the network. This lowers the difficulty until the cost of mining again approaches the price.


Unless the Price Recovers Sharply, the Hashrate Will Fall, Then the Difficulty

The bitcoin price is now down more than 50% from its 2020 highs. This means that miners’ income has lost more than 50% of its purchasing power. The highest cost producers will now be unprofitable and some will even be cashflow negative. When miners turn cashflow negative they will turn off their gear and hashrate will fall.

This will cause blocks to be produced more slowly than before. In turn, this means that the next time the difficulty is adjusted, it will be lowered, such that the remaining hashrate finds blocks at the same frequency as before.

Whichever miners are able to remain cashflow positive during this transition period will end up having their cost of mining reduced. After the difficulty adjusts downwards, a new balance is found where the remaining miners find a higher percentage of blocks than they used to (aka, their bitcoin income is increased, making up for the loss of reduced prices per bitcoin sold).

Bitcoin-denominated income therefore flows from inefficient high-cost producers to efficient low-cost producers, Bitcoin as a system will draw less power than it did before, and the average power it draws will be lower cost (aka, less useful for society) than the average power it drew before.

The main overall effect is that the cost of writing (and rewriting) the chain decreases. But no one really knows how much chain-writing cost is enough, and that topic is way outside of the realm of what we can fit in this article (if you’re interested we can recommend starting here, here and here).

‘Mining Death Spirals’ Exist Only in Theory, not in Practice

In price environments like these you will hear countless uninformed pundits peddle the ‘mining death spiral’ concern. The weak form of the argument generally goes like this:

If bitcoin prices fall, mining will become unprofitable causing the hashrate to drop. This will grind the network to a halt since no new blocks are mined. This in turn will cause price to drop further causing more miners to shut down until no one is left mining and the price hits zero.

From reading this article you’ll already understand why that form of the argument is absolute nonsense (and also the reason we’ve never observed it in practice even though we’ve had two halvings already and the bitcoin price has fallen by more than 50% on several occasions): the mining difficulty adjustment.

However, there is an edge case in which a mining death spiral could conceivably happen. The steel man form of the argument goes like this:

If,

  1. At the worst possible time (that is, exactly after a difficulty adjustment), the price of bitcoin drops by an extreme amount (let’s call it 99%), and;
  2. Very few or none of the miners believe it will ever come back up, and;
  3. Miners are physically able to almost immediately shut most (let’s call this 99% as well) mining gear off.

The result would be that the time it would take to find 2016 blocks for the next difficulty adjustment would be 100 times longer than normal (that is, 200 weeks, or almost 4 years). Moreover, the maximum difficulty adjustment is 4x in each direction, so the next difficulty adjustment would take another 50 weeks, then another 12.5 weeks, then another 4 until block times were normalised.

In such a scenario you can imagine that the price would falter (although hasn’t it already?) and that the future of the system could be called into question.

In real life though, markets don’t move like that. And if they did, a permanent 99% (or whatever) price drop would likely be a signal that something else was fundamentally wrong with the system itself.

On top of that there are operational concerns on the part of miners that prevent shutdowns on such rapid timelines. Powering down a several hundred megawatt mine is not a matter of pulling a socket plug––you would risk severely damaging the local grid. Moreover, many miners have offtake agreements that mandate that they continue their draw for as long as they are able to pay their contracted bills.

The point is: even when bitcoin prices significantly fall (which happens pretty much every year) or the mining reward is halved (which happens at predetermined time intervals), the physical and operational realities of the mining network are such that drawdowns in hashrate take time. In practice, hashrate reductions are therefore always smoothly caught by the dynamic difficulty adjustment and block frequencies never get anywhere near ‘crisis levels’ (whatever that even means).


The Network Was Designed to Handle These Exact Situations

If you’re worried that price drops or halvings will have long-term adverse effects on the operational workings of Bitcoin consider the following: This is the umpteenth time the price of bitcoin has seen a dramatic pullback––it happens pretty much every year at some point or another––and this is the third halving of the block reward.

Because of the design choices we’ve explained above the mining network has never failed to produce blocks. The difficulty has reset downwards many times––sometimes dramatically as the result of a pullback in price (the November/December 2018 is an excellent example to study), but never has the network ground to a halt or even come anywhere close to it.

Source: CoinShares Research


Bitcoin as a monetary system is extremely robust. No matter where the price goes, block generation times are coldly and unemotionally kept at their preset levels by computers, not humans. There is no price level that could cause Bitcoin’s emission rate to increase. When the dust settles on the current financial crisis, the Bitcoin monetary system will have created exactly as many bitcoins as originally intended.

The same is not true of its government competitors.




Disclosure

This material has been prepared by CoinShares and its affiliates for educational and informational purposes only and it is not intended to be relied upon as an offer or a recommendation, offer or solicitation to buy or sell a security nor is it to be construed as investment advice. Predictions, opinions and other information are expressed at the date of publication and are subject to change as circumstances vary. This information has been developed internally and/or obtained from third party sources believed to be reliable; however, no representation or warranty, express or implied, is made as to the accuracy, reliability, or completeness of such information. To the extent permitted by law, we do not accept or assume any liability, responsibility or duty of care for any use of or reliance on this information. Past performance is not a reliable indicator of future performance.

The CoinShares Astronaut is a trademark and service mark of CoinShares (Holdings) Limited.

Copyright © 2020 CoinShares Group, All rights reserved.

Newsletter

Sign up for our monthly newsletter

Subscribe

Our latest insights & research. Never spam.