Hypothetic question regarding network security after last halving

[Kryptowerk]

Recently I had a discussion with a friend interested in- but not convinced by- the principles of Bitcoin.
You know how it goes "blabla I don't really see the longterm value of Bitcoin" - "blabla, currently store of value, immutable, can scale with lightning to become widely used for payments, predictible outcomes, no centralized control, safe-heaven against institutional/gov. control bla".
Some back and forth.

He then asked me this question, which I wasn't entirely sure how to asnwer. IF Bitcoin just stays a store-of-vaue/digital-gold type of thing, because it has many advantages in comparison to physical store-of-values but never is mainly used for payments: How can the network incentive to keep running lots and lots of expensive mining hardware with high-energy-consumption (lets forget about the possibility for other proof-of-work algos and free energy solutions like fusion power etc for this example) be maintained? If the network is only used every once in a while for a transfer of value from one entity to another?

We are assuming block-reward is 0 now and the only reward is coming from network fees. (after last halving)

To put it in other words: Wouldn't be that amount coming from fees be possibly too small to keep up a healthy hashrate and therefore weaken the network / breaking the current equilibrium of hashing-effort vs. block-reward-payout? Assuming Bitcoin transactions are very low in volume, because it's not used for payments, just as a store of value/digital-gold?

Not entirely sure if this belongs in technical discussion, my apologies in advance if this is misplaced.

[1714716481]

1714716481

[1714716481]

1714716481

[1714716481]

1714716481

[Charles-Tim]

Some people can ask when bitcoin mining reward was 50 BTC, that if bitcoin mining reward halves to 6.25 BTC, that the network would collapse. But what happened? Increase in hash rate and mining difficulty.

If mining is not profitable for some miners, they will shutdown their miners and the difficulty will reduce while other miners that remain will make more money from the transaction fee.

But if bitcoin adoption continues and the price continue to increase in long term as it has been increasing, expect all-time-high in mining hash rate.

[Kryptowerk]

Some people can ask when bitcoin mining reward was 50 BTC, that if bitcoin mining reward halves to 6.25 BTC, that the network would collapse. But what happened? Increase in hash rate and mining difficulty.

If mining is not profitable for some miners, they will shutdown their miners and the difficulty will reduce while other miners that remain will make more money from the transaction fee.

But if bitcoin adoption continues and the price continue to increase in long term as it has been increasing, expect all-time-high in mining hash rate.

I think you are missing the point, there is huge difference of -50% reward, which automatically results in *parts* of the mining hardware turning unprofitable (depending on their energy consumptiond and associated price), assuming constant price of BTC. This will be balanced within an equilibrium range of profitability vs hardware+energy-costs.
However, if it the reduction is 100% and all that's left is transaction rewards, this seems like an unpredictable jump that could cause some huge trouble. It increases variance (miners are depending on transactions alone now) - If tx-fees and tx-amount are too low, this could result in a massive drop of hashrate and therefore possibly open up a relatively cheap 51% from a malicious actor.
That's at least what we were wondering about.
It's still quite some time to go (I think 2140the last BTC will be mind), but an interesting and maybe important thing to think about.

[lassdas]

.. (I think 2140the last BTC will be mind),...

No need to wait that long
and it won't be an unpredictable jump that suddenly happens overnight.

Looking at the numbers the current fees are somewhere around 0.1-0.2ish BTC per block,
checking the "halving roadmap" in this thread we see that in ~20years, the block-reward will be around that same range too.
Currently the block-reward is the main-dish and the fees are just a bonus, from 2040-2044 on it's the other way around.

And every miner knows (or should) this already and has enough time to either adapt or slowly leave the playing field to make room for others.

Besides, I'm sadly not around anymore in 100+years, but I would be mining at a loss if that keeps the network going and I'm just a random nobody,
why wouldn't some big player like an exchange, payment provider, or whatever bitcoin-business spend some of its profits on mining-activities if it keeps the cash-cow poop?
If there's another income-stream dependent on the bitcoin-network to run smoothly, mining itself doesn't need to be profitable.


Anyway, to predict what might or might not happen in the future is as hard as always.

[thecodebear]

OP, first off, why are you assuming transaction volume would be low??

Blocks are already full and have been full for years even though it is barely ever used for payments right now. You seem to be saying "what if transactions become less frequent", but obviously that won't happen. Blocks will stay full as Bitcoin becomes more popular of course. And transaction fees will go up.

Now hopefully there will be further improvements to make transactions a bit more efficient, or at least certain types of transactions more efficient, and maybe even ways to cram more data in, like the way Segwit did.

Also it's not like this will suddenly happen one day. The whole last halving cycle before Bitcoin hits its supply limit the block rewards will only be 1 satoshi. The block reward will be negligible for decades before it finally stops. And starting from now it will take decades for that block reward to become negligible. If you think about the block rewards halving vs the increase in price, by around 2050 the halvings will likely have outpaced the price appreciation starting from now, so by around 2050 it's likely that miners will be receiving less money (in terms of USD) than they do today. So it'll take a bunch of years until the halvings even start to effect the mining industry, and it'll be like 90 years from that point until there are no more satoshis being rewarded, and all during this time the tx fees will have to gradually make up the difference. So it's not like one day mining will be fine and the next day there will be a huge issue. This means if this does cause a problem there will literally be decades to deal with it.

Realize that the Bitcoin network already has global-level security and is the most secure network in human history. So it's not like we NEED the mining network to keep growing, we just need it to stay decentralized. Also realize that mining machines grow more efficient over time, so hash rate will continue to increase even without more energy (cost of mining) used. Also Bitcoin mining is still a very new industry and in the coming decades as it becomes accepted more and matures it will naturally go to the most efficient places (like running on unused power at renewable power stations) or running at purpose-built renewable power stations where they are generating energy at nearly zero cost after initial startup costs, rather than like paying on-grid prices like most miners do today.

Also the most obvious point - mining difficulty can adjust down for any negative chance in hash power going into mining, so even if hash rate peaks at some point and goes down after that until settling at a tx-fee-only equilibrium, the system will adjust for that and it'll just mean the most efficient miners stay in business. Only thing that matters is it stays decentralized among those most efficient miners, which seems pretty likely since you can mine from anywhere in the world as long as you have access to cheap power.


In summary:
1. Blocks will obviously stay full so its not like there will be fewer transactions or less competition for getting transactions in blocks

2. Potential for further improvements in the bitcoin protocol in the coming decades to allow more transactions to be crammed into blocks

3. Transaction fees will go up as bitcoin use grows

4. The effects of the mining reward becoming negligible will happen over decades, not all at once, so if there is a problem it can be dealt with over decades

5. Mining machines will become more efficient to use less power and naturally miners will settle towards using only the cheapest electricity, bringing down the cost of mining per unit hash rate

6. Hash rate doesn't need to go up forever, Bitcoin is already the most secure network in human history, we should certainly expect at some point in the next few decades hash rate will peak (my guess would be sometime in the mid-to-late 2040s) and then settle somewhere at a long term equilibrium, the halvings will eventually cause hash rate to peak and the difficulty adjustment will allow it to settle at equilibrium somewhere lower than that peak once the block rewards of new bitcoin are a negligible part of the miners income

7. This is all perfectly fine, the only thing that matters is that mining stays decentralized once this equilibrium is hit when miners are making their income just from tx fees.

[thecodebear]

Some people can ask when bitcoin mining reward was 50 BTC, that if bitcoin mining reward halves to 6.25 BTC, that the network would collapse. But what happened? Increase in hash rate and mining difficulty.

If mining is not profitable for some miners, they will shutdown their miners and the difficulty will reduce while other miners that remain will make more money from the transaction fee.

But if bitcoin adoption continues and the price continue to increase in long term as it has been increasing, expect all-time-high in mining hash rate.

I think you are missing the point, there is huge difference of -50% reward, which automatically results in *parts* of the mining hardware turning unprofitable (depending on their energy consumptiond and associated price), assuming constant price of BTC. This will be balanced within an equilibrium range of profitability vs hardware+energy-costs.
However, if it the reduction is 100% and all that's left is transaction rewards, this seems like an unpredictable jump that could cause some huge trouble. It increases variance (miners are depending on transactions alone now) - If tx-fees and tx-amount are too low, this could result in a massive drop of hashrate and therefore possibly open up a relatively cheap 51% from a malicious actor.
That's at least what we were wondering about.
It's still quite some time to go (I think 2140the last BTC will be mind), but an interesting and maybe important thing to think about.

Why are you talking about a "100%" and "unpredictable jump" Nothing in the halvings is unpredictable. We know at exactly every block number what the mining reward will be. There is no unpredictable jump.

And as far as the 100% drop...yeah it's literally a drop from 1 satoshi to 0 satoshis around 2140. By the time that happens miners income will have been entirely dependent on transactions fees for many decades. It seems like you think that all of a sudden one day the mining reward will go from entirely supporting miners to being zero. That's not how this works. The halvings mean that mining rewards drop gradually over the course of many decades. Even if the price of a bitcoin is $100,000,000 on the day the mining reward stops, that means miners will be receiving $1 less per block after the final halving lol. The mining industry will find an equilibrium over many decades. There is no worry about one day a 51% attack will be easy because mining rewards suddenly disappeared, because mining rewards never suddenly disappear. Over decades tx fees will gradually come to support more and more of mining until they are supporting all of mining, and the hash power will settle at a place where mining is profitable with only tx fees, and the mining difficulty adjustment allows that to happen.

[Sha256explorer]

if in 2140 (and before)  the price of Bitcoin will be high and the fees will be high, will the Lightning network always be cheap at that point?

[mikeywith]

Bitcoin's security is directly related to its value, the more value it has, the more people can afford to mine it, and the number of satoshis mined is irrelevant, less block rewards = less supply which in theory should result in more value to be extracted by mining BTC since it's value has increased, in other words:

100k BTC 30k BTC and a block reward of 3.125 BTC is better than 30k BTC and block reward of 6.25 BTC, in other words, if you want BTC to be as secured as it's today with only block fees that amount of fees needs to 6.25 BTC at 30k BTC, if the fees are only 0.25 BTCinstead, then the price of BTC has to be $750,000.

So no block rewards and BTC worth $750,000 is exactly the same as 6.25 BTC block rewards and BTC worth $30,000.

Something else to keep in mind is that if BTC keeps growing the same way, those who have value at stake will be willing to mine it at a loss just to protect their BTC.

With that said it would be best if we make sure that blocks are always full or near full, we don't know how is that going to be achieved or how will people view and use Bitcoin in 70 years, the good thing about this is the fact that it will unfold slowly, so no urgent measures have to be taken, when the time comes, the next generation could be reading our comments and laugh at how we got everything wrong.

[pooya87]

Something that people using this argument involving halving forget is that each halving the reward reduction is smaller than the previous one. The first halving the block reward dropped by 25 bitcoins whereas the upcoming halving the reward will only drop by 3.125 and when the reward reduced by 25 bitcoins, the hashrate only grew, there is no reason to argue that when the reward reduces less the hashrate would drop.

In a couple of more halvings the before and after reward won't be that different.

[HeRetiK]

He then asked me this question, which I wasn't entirely sure how to asnwer. IF Bitcoin just stays a store-of-vaue/digital-gold type of thing, because it has many advantages in comparison to physical store-of-values but never is mainly used for payments: How can the network incentive to keep running lots and lots of expensive mining hardware with high-energy-consumption (lets forget about the possibility for other proof-of-work algos and free energy solutions like fusion power etc for this example) be maintained? If the network is only used every once in a while for a transfer of value from one entity to another?

Low fees lead to Bitcoin being used as means of payments. High fees lead to a shift of Bitcoin as store of value that gets transacted rarely. What we currently see, is something inbetween. And I believe it will remain that way, as it seems that we always reach an equilibrium between the fee market and how people use Bitcoin, similar to how we always reach an equilibrium of profitable mining.

if in 2140 (and before)  the price of Bitcoin will be high and the fees will be high, will the Lightning network always be cheap at that point?

The economics of LN fees are different from the economics of mining fees, enabling Lightning transactions to remain much cheaper than on-chain transactions. Discussing that in detail would get off-topic though.

[Synchronice]

I think, bitcoin wasn't meant for very long-term usage with it's current form, that's why we have an ability to fork bitcoin. Personally, I think that from current situation, it will be a good deal if we increase bitcoin block size and we already have a fork of it called Bitcoin Cash but somehow people hate this coin on this forum, maybe it has something to do with Roger Ver, idk but I still think that block-size increase is a future-wise decision because definitely Bitcoin price will be very big and increases block size will give users possibility to spend transactions with low transaction fees but high number of them will generate good profit for miners, enough to continue their operation.

But it's impossible to predict what will happen when the last block will be mined because of very unpredictable future. Just think about it, in 1860, probably no one would imagine any way of communication from place A to place B on long distance but In 1876 we invented first telephone. Would anyone imagine wireless communication from A country to B country? No, but in 1895 Marconi succeed in receiving Morse code on a radio wave that let to the development of the first handheld cellular phone that was invented in 1973. And today, in 2023, we have smartphones that not only give us possibility to contact from any place to any place, we can take a picture, record a video, voice, play video games, do complex tasks.

The last bitcoin will be mined in 2140. We have 117 years left, this is such a huge time and the world develops so faster, nothing is predictable to my mind.

[stompix]

Some people can ask when bitcoin mining reward was 50 BTC, that if bitcoin mining reward halves to 6.25 BTC, that the network would collapse. But what happened? Increase in hash rate and mining difficulty.
If mining is not profitable for some miners, they will shutdown their miners and the difficulty will reduce while other miners that remain will make more money from the transaction fee.

Mining didn't collapse because the actual reward per block kept going up.
How will the reward keep going up when there is no block subsidy but only fees unless you force people to pay more?

Not to mention a different thing, if the hash rate for example will be the same even as the halving pass by, as the price will offset it, you will still face a problem, the same hashrate will have to defend a blockchain that is worth 2x more, security has to be measured on how much it takes to overcome it but also how much it protects. 300Exash and 3 million machines sound nice right now, but if in 3 halvings this will still be the same but with the market cap no longer at 600 billion but 5 trillion, will it mean the network is just as secure?

3. Transaction fees will go up as bitcoin use grows

So if Bitcoin reaches $ 1mil you will have no problem paying $30 for a simple 1 input 1 output tx and you think people will still be using it in their daily lives?
Hasn't the drama of the Ordinals and people screaming like mad that there is a terrorist attack because they have to pay 4sat/b shown how people react ?

5. Mining machines will become more efficient to use less power and naturally miners will settle towards using only the cheapest electricity, bringing down the cost of mining per unit hash rate

No matter how efficient the machines are the cost to attack the network will move the same way as the efficiency of the machines.
12 years ago you could attack it with a single s19 but because it didn't exist you had to spend millions in FPGA, so it's not the efficiency of the machine is the cost of the 51% in current Th/$ that keeps the network secure.

[death69]

Alright, buddy, let me break it down for you – if Bitcoin's still kicking in a century, we'll probably be coexisting with a robotic-alien hybrid society.
No joke, though, the query about incentivizing Bitcoin mining in the future is legit. As the network expands and transforms, so must the rewards that fuel it.

From alternative proof-of-work schemes to fusion-driven mining beasts (no kidding!), the potential is vast. Even if Bitcoin morphs into a digital gold reserve, its intrinsic value will continue to attract believers. And where faith exists, power resides. So let's not fret over the distant future. We've got more immediate concerns, like surviving the robot-alien extravaganza.

[hatshepsut93]

If we want Bitcoin to be secured by a similar amount of work in the future, it means the fees will have to compensate for the halvenings. And Bitcoin capacity is so limited that even with only store of value use the fees will skyrocket if it gets a little more adoption. We see it during bull markets how fees become $10-20 per tx just because so many people want to send their coins to an exchange.

But I wouldn't exclude a possibility that in the future there will be no other choice than to return block rewards and make Bitcoin inflationary, because otherwise the network security will be too low.

[Kryptowerk]

OP, first off, why are you assuming transaction volume would be low??

Blocks are already full and have been full for years even though it is barely ever used for payments right now. You seem to be saying "what if transactions become less frequent", but obviously that won't happen. Blocks will stay full as Bitcoin becomes more popular of course. And transaction fees will go up.

Now hopefully there will be further improvements to make transactions a bit more efficient, or at least certain types of transactions more efficient, and maybe even ways to cram more data in, like the way Segwit did.

Also it's not like this will suddenly happen one day. The whole last halving cycle before Bitcoin hits its supply limit the block rewards will only be 1 satoshi. The block reward will be negligible for decades before it finally stops. And starting from now it will take decades for that block reward to become negligible. If you think about the block rewards halving vs the increase in price, by around 2050 the halvings will likely have outpaced the price appreciation starting from now, so by around 2050 it's likely that miners will be receiving less money (in terms of USD) than they do today. So it'll take a bunch of years until the halvings even start to effect the mining industry, and it'll be like 90 years from that point until there are no more satoshis being rewarded, and all during this time the tx fees will have to gradually make up the difference. So it's not like one day mining will be fine and the next day there will be a huge issue. This means if this does cause a problem there will literally be decades to deal with it.

Realize that the Bitcoin network already has global-level security and is the most secure network in human history. So it's not like we NEED the mining network to keep growing, we just need it to stay decentralized. Also realize that mining machines grow more efficient over time, so hash rate will continue to increase even without more energy (cost of mining) used. Also Bitcoin mining is still a very new industry and in the coming decades as it becomes accepted more and matures it will naturally go to the most efficient places (like running on unused power at renewable power stations) or running at purpose-built renewable power stations where they are generating energy at nearly zero cost after initial startup costs, rather than like paying on-grid prices like most miners do today.

Also the most obvious point - mining difficulty can adjust down for any negative chance in hash power going into mining, so even if hash rate peaks at some point and goes down after that until settling at a tx-fee-only equilibrium, the system will adjust for that and it'll just mean the most efficient miners stay in business. Only thing that matters is it stays decentralized among those most efficient miners, which seems pretty likely since you can mine from anywhere in the world as long as you have access to cheap power.


In summary:
1. Blocks will obviously stay full so its not like there will be fewer transactions or less competition for getting transactions in blocks

2. Potential for further improvements in the bitcoin protocol in the coming decades to allow more transactions to be crammed into blocks

3. Transaction fees will go up as bitcoin use grows

4. The effects of the mining reward becoming negligible will happen over decades, not all at once, so if there is a problem it can be dealt with over decades

5. Mining machines will become more efficient to use less power and naturally miners will settle towards using only the cheapest electricity, bringing down the cost of mining per unit hash rate

6. Hash rate doesn't need to go up forever, Bitcoin is already the most secure network in human history, we should certainly expect at some point in the next few decades hash rate will peak (my guess would be sometime in the mid-to-late 2040s) and then settle somewhere at a long term equilibrium, the halvings will eventually cause hash rate to peak and the difficulty adjustment will allow it to settle at equilibrium somewhere lower than that peak once the block rewards of new bitcoin are a negligible part of the miners income

7. This is all perfectly fine, the only thing that matters is that mining stays decentralized once this equilibrium is hit when miners are making their income just from tx fees.

These were some very nice elaborations, exactly what I was hoping for.
Let me clarify a little bit, I didn't really assume tx-volume would be low by then, I was mostly playing "devils-advocate" (more like friend-of-mine advocate), because he was wondering what would happen in that specific case (low tx due to BTC just being a store of value).
Have a great weekend!

Alright, buddy, let me break it down for you – if Bitcoin's still kicking in a century, we'll probably be coexisting with a robotic-alien hybrid society.
No joke, though, the query about incentivizing Bitcoin mining in the future is legit. As the network expands and transforms, so must the rewards that fuel it.

From alternative proof-of-work schemes to fusion-driven mining beasts (no kidding!), the potential is vast. Even if Bitcoin morphs into a digital gold reserve, its intrinsic value will continue to attract believers. And where faith exists, power resides. So let's not fret over the distant future. We've got more immediate concerns, like surviving the robot-alien extravaganza.

Haha, I feel ya. Yes, obviously there are an enormous amount of challanges to overcome. I still enjoy to talk about some very hypothetical things, especially in this case, where, as long as there is no major hardfork, things on the Bitcoin side are very deterministic and thus allow for this sort of discussion without it turning into a total cloud-castle.

[franky1]

Blocks are already full and have been full for years even though it is barely ever used for payments right now. You seem to be saying "what if transactions become less frequent", but obviously that won't happen. Blocks will stay full as Bitcoin becomes more popular of course. And transaction fees will go up.

Now hopefully there will be further improvements to make transactions a bit more efficient, or at least certain types of transactions more efficient, and maybe even ways to cram more data in, like the way Segwit did.

first fail of economics is thinking people will continue PAYING "MORE FEES"
the solution is MORE TRANSACTIONS that combine to create MORE TOTAL INCOME for pools

thats the big difference

cramming more USELESS data is not the solution. if less users are making transactions because only 5 people are filling blocks becasue the 5 filling a block are are mining pool block creator paying itself by filling its own block with its own transactions. is a FAILURE of economics and monetary policy

again the solution is more transactions per block to allow more users who end up paying less fee's individually but it all combines to give higher total for the pool to be rewarded. thats how you secure the network and keep good network utility