Tail emission ideas that retain the 21 million limit

[stwenhao]

https://en.bitcoin.it/wiki/Prohibited_changes

These changes require the consent of every bitcoin-holder:

• Increasing the total number of issued bitcoins beyond 21 million. Precision may be increased, but proportions must be unchanged.

Which means, that it should be optional. It is technically possible to make a chain, where each user can decide with signatures, if he wants to support that change or not. But, as I said before: burning coins is easier than making them out of thin air, in a backward-compatible way. Those, who disagree, will simply keep using the old version.

And also, the most likely outcome, is just some kind of fork, where there will be two coins: the original one, and the one with tail supply. And I guess it will be hard to convince all users, that the coin with tail supply is the real Bitcoin, and not just yet another altcoin. Because if tail supply supporters would want to avoid any forks, then they would need to trace the chain with the heaviest Proof of Work, and there are many reasons, why tail supply version may not reach hashrate majority.

Which also means, that tail supply supporters should be prepared to work within existing rules, because there are much more ways to even accidentally jump into the altcoin land, than they are to stay within Bitcoin (which is also, why there are so many altcoins: it is much easier to make yet another altcoin, than it is to improve Bitcoin, and stay compatible).

Another hint is to start from test networks: there are other prohibited changes, like network centralization. This assumption was broken in signet. Which means, that the whole idea will have much more chances to succeed, if it will be done as a test network first. And maybe it will even benefit from staying as a testnet, because weaker coins are for spending, and stronger coins are for saving.

[SilverCryptoBullet]

https://en.bitcoin.it/wiki/Prohibited_changes

These changes require the consent of every bitcoin-holder:

• Increasing the total number of issued bitcoins beyond 21 million. Precision may be increased, but proportions must be unchanged.

Which means, that it should be optional. It is technically possible to make a chain, where each user can decide with signatures, if he wants to support that change or not. But, as I said before: burning coins is easier than making them out of thin air, in a backward-compatible way. Those, who disagree, will simply keep using the old version.

Although this can be done technically, nobody want this unprecedented increase of Bitcoin total supply from its initial cap at 21M coins. Having a finite total supply is one of biggest strengths of Bitcoin compares to overwhelming thousands of shitcoins on the market. Latecomers might want to increase the total supply but as late comers, they have no power to change anything while early comers don't support this idea.

If this proposal is done and total supply increased, it does not help late comers to benefit from this milestone in Bitcoin history, and it only brings nightmare to all holders who are either early adopters or newbies in the market.

If you need any example, see how the Purchasing Power of US. dollar and other fiat currencies drop dramatically over time as consequences of inflation.
https://www.visualcapitalist.com/purchasing-power-of-the-u-s-dollar-over-time/

In contrast, purchasing power of 1 bitcoin or 1 satoshi has increases paraoblically with time.
https://charts.bitbo.io/satoshi-per-dollar/

Extra source for reading.
https://decrypt.co/34876/why-is-bitcoins-supply-limit-set-to-21-million
https://blog.lopp.net/how-is-the-21-million-bitcoin-cap-defined-and-enforced/
https://river.com/learn/can-bitcoins-hard-cap-of-21-million-be-changed/ - this article is insightful.

[tromp]

Having a finite total supply is one of biggest strengths of Bitcoin compares to overwhelming thousands of shitcoins on the market.

It's not really. Being predictable and disinflationary are great strengths, but neither of those require a finite supply [1].

If you need any example, see how the Purchasing Power of US. dollar and other fiat currencies drop dramatically over time as consequences of inflation.

Fiat is a bad example, since unlike most cryptocurrencies, it's not even disinflationary.

[1] https://tromp.github.io/blog/2020/12/20/soft-supply

[stwenhao]

It doesn’t take much to change a finite supply into an infinite one.

Only if you want to make a hard-fork, which will quickly put you into an altcoin land.

What if that final drop never happened?

Blocks with additional coins would be considered invalid by old nodes, and would turn all users of a new version into an altcoin.

Would that make make Bitcoin any less hard a currency?

Note that in the current consensus, it is possible to burn coins. Technically, a coin can be made in a way, where all coins will always be spendable, but Bitcoin is not such coin.

So what makes a currency hard, if not a capped supply?

Mining. Proof of Work is what makes it hard, because then, changing things require consuming real energy, and this is what makes it hard to produce coins out of thin air. A good example of that can be seen in test networks, where testnet3 and testnet4 have a lot of chainwork, and are traded for real BTCs (because producing test coins requires mining, and it is costly), while signet, with quite low chainwork, is worth literally zero (also because it is centrally controlled, and developers can halt it at any time, by refusing to sign next blocks).

How low should inflation be to be considered negligible?

As long as people can burn coins, it doesn't matter that much. On top of a chain with tail supply, users can agree to burn their coins collectively, and to keep the supply limited, no matter how many coins will be produced. I guess if some testnet would really have doublings, instead of halvings, and if eventually people would get 21 million coins per block forever, then some people will opt into coin burning, just to raise the value of their coins. Also, they will introduce "ordinals", "rare satoshis", "NFTs", and other kinds of things, which would be worth more, than other coins around. And then, "1 satoshi != 1 satoshi" for some people, because things are worthy, if you put some work into them (the most trivial example is mining, but it is not limited to just that).

Edit: To better see, why Proof of Work is important, consider this address: tb1qndpzf7522jtn7mfstwjqcn55rrlqxpzmqadyv3h4mgtk2m23xhtqa40qwx. How much is it worth? You can clearly see "50 tBTC4" here, but if you would want to move it, then it would require you to make a valid signature for the private key equal to one, and grind it to take below 50 bytes. Which means, that you would need to check 2^88 hashes, to find some matching s-value, if you would use R-value as a half of the generator, and focus on grinding s-value with SHA-256.

So, coins from this puzzle are probably worth more, than the amount of test coins would suggest, because everyone, who would want to touch them, would have to put a lot of Proof of Work, to sweep them, even though the private key is known, and equal to one. There are no shortcuts I know of, so I also don't know, how to sweep them faster, than by making a valid transaction, and grinding the solution.

[takuma sato]

It doesn’t take much to change a finite supply into an infinite one.

Only if you want to make a hard-fork, which will quickly put you into an altcoin land.

I would agree typically with this but with one exception: The quantum computer threat is very real in the long term, and as far as I know this may require a hardfork, and here I fear that people are going to want to stuff on the hardfork every idea under the sun, and since their idea does not get agreed by the majority there will be infighting. Im just hoping rationale may remain and we will just agree to something makes the most sense specifically for QC threat defense and that's about it. But some people may bring for instance this subject of increasing the tail emission, even if some novel ideal where it would somehow respect the 21 million limit which I don't see how. But that is the only scenario where I see an actual hardfork pick up traction due the QC threat. If we can get QC protection with no hardfork then that would be ideal.

[Satofan44]

It doesn’t take much to change a finite supply into an infinite one.

Only if you want to make a hard-fork, which will quickly put you into an altcoin land.

I would agree typically with this but with one exception: The quantum computer threat is very real in the long term, and as far as I know this may require a hardfork, and here I fear that people are going to want to stuff on the hardfork every idea under the sun, and since their idea does not get agreed by the majority there will be infighting. Im just hoping rationale may remain and we will just agree to something makes the most sense specifically for QC threat defense and that's about it. But some people may bring for instance this subject of increasing the tail emission, even if some novel ideal where it would somehow respect the 21 million limit which I don't see how. But that is the only scenario where I see an actual hardfork pick up traction due the QC threat. If we can get QC protection with no hardfork then that would be ideal.

I don't think that is what is going to happen, unless of course the people that propose the hard fork are idiots. If or when the threat of quantum computers are real, then a changes must be made to ensure the safety of Bitcoin. If a hard fork is needed, then it must pass as quickly as reasonably possible or as needed and with the least amount of contention. If you have Bitcoin's best interest at heart, you will not stuff anything contentious in the same hard fork because this will guarantee its rejection. For me, if we do have to do a hard fork it would be probably to only include non contentious changes that are needed to clean up some technical debt and other issues. Anyhow, I think that we can probably get away without a hard fork but this topic is still developing.

Having a finite total supply is one of biggest strengths of Bitcoin compares to overwhelming thousands of shitcoins on the market.

It's not really. Being predictable and disinflationary are great strengths, but neither of those require a finite supply [1].

You are making assumptions and judgements for other people, do not do this. Everyone gets to decide what is Bitcoin's biggest strength for themselves, and which features provide the most value to them. Everyone that I know that has Bitcoin is pretty much strict on the finite total supply, and I am very sure that a lot of people are in the same position. Slipper slope danger anyway, today you're talking about removing the finite supply and tomorrow you are going to talk about yields and staking.  

[mikeywith]

If the protocol prevented them from grabbing that "burnt" part of the transaction fee, then they would not have this option. Or maybe you refer to the possibility that miners could reject that soft/hardfork? I think in this case everything would depend on the signalled support from economic nodes.

Then you'll have to mine your own blocks. Just as miners can't modify the block subsidy to give themselves double the reward, you also can't just rewrite the rules to take half the fees from them. You do realize that miners are the ones who actually write the blockchain. Try telling them, "Hey, we're going to pay you less now, please opt in to this new fork." Yeah... that's not going to happen, and without miners signalling support for this, the fork will be very short-lived.

The proposal would probably lead to a more stable hashrate even in periods with low onchain activity, so miners should support the proposal also for their own good to make it more predictable.

How is that for their own good? You're talking about rewarding miners in the future at the expense of miners working today. The logic of "suffer today so others can rest tomorrow" doesn't hold up from a miner's perspective. You're asking current operators to take a guaranteed cut now in exchange for a vague promise of “predictability” later, which isn't how capital-intensive, high-risk operations like mining work.

The reality is this: BTC security depends entirely on its value. It's a self-securing protocol that has proven to work for many years, and I don't see why it wouldn't continue working long into the future. Over time, as block rewards decline, mining will be subsidized by Bitcoin holders one way or another.

When large corporations have a significant stake in BTC, they'll want to support the network's security. Even the average user would likely be willing to contribute something to protect the value of their holdings if block rewards become too small to sustain the ecosystem. This kind of subsidy could be proportional to each person's holdings -- like paying a few dollars a year to keep your $10,000 safe. I don't think most people would object.

The technical mechanisms for how this would be implemented can be discussed when the time comes. But to me, it seems like the only viable long-term path forward -- if it ever becomes necessary. You might call it a tax, but it's really not. Just like you pay fees to move your coins, you may end up paying fees to store them securely.

[stwenhao]

The quantum computer threat is very real in the long term, and as far as I know this may require a hardfork

It can be done as a soft-fork. There are other things, which are more likely to result in a hard-fork, for example running out of timestamps in 2038 (or later, because block time is sometimes handled as int32, and sometimes as uint32).

I fear that people are going to want to stuff on the hardfork every idea under the sun

Of course people will want to do that. But as usual, when you want to reach consensus even on soft-forks, there are many things, which are not implemented, because of lack of agreement. So, I guess people will want to do a lot of things during hard-fork, but consensus will be reached only for very minimalistic changes, and only that will be implemented in practice (and also it will be very limited, to imitate soft-fork as close as possible, and have as little points of incompatibility, as possible).

even if some novel ideal where it would somehow respect the 21 million limit which I don't see how.

It is very simple: as long as total supply is never infinite, by checking the supply in block number N, you can do the mapping between finite and infinite model. Which means, that if you change proportions accordingly, then you can reach the same economy in both. The difference can be simplified into this: in finite supply, users can see single satoshis, taken out of their accounts, and allocated for future mining rewards. In infinite supply, exactly the same thing is done, but coin amounts are preserved, like in fiat currencies, but they are worth less, than they were in the past, according to the level of created inflation.

If we can get QC protection with no hardfork then that would be ideal.

Of course we can. As long as only ECDSA is broken, and not SHA-256, every OP_CHECKSIG could require more conditions, than it requires today.

[markm]

Having read through this thread, it seems to me to provide a reason for thinking that if having any second third etc layers etc on top of bitcoin it would be useful for them to use merged mining, so that even if some day some of them become more profitable to mine than bitcoin itself nonetheless miners could earn more by including bitcoin itself in their merge than by leaving it out.

Some of the merged mined coins also already have endless tail-emission, e.g. GRouPcoin which just issues 50 coins per block forever.


-MarkM-

[Wind_FURY]

The first proposal would require some pretty invasive changes to consensus (hard fork), and I'm not sure that the advantages would outweigh the disadvantages. But either way - when block subsidy goes to zero, wouldn't miners be incentivised to fill their blocks? If we assume full or mostly full blocks and a minimum fee rate (e.g. 1 sat/vB), there will still be a baseline predictable block reward.

 👍

Plus probably some unnecessary complexity, and possibly open the network to exploitation of the new system?

The second proposal doesn't have to be a second "official token". Because "official" probably means endorsed by the Core Developers? We want them to be shitcoiners? That's not going to happen ser.

The Mining Cartel will probably start some sort of sidechain by themselves, or another blockchain with unlimited supply entirely that's merge-mined with Bitcoin to sustain themselves.

[z5k_alt]

Then you'll have to mine your own blocks. Just as miners can't modify the block subsidy to give themselves double the reward, you also can't just rewrite the rules to take half the fees from them.

I think you overestimate the miners' power in the game. If the Bitcoin users and holders think that the version of Bitcoin is superior to the version without that burning (e.g. because its security is superior), then there will be enough miners happy to mine the new protocol. And if they don't want to support a softfork you can threaten with a hard fork. The Bitcoin users have always the last word, as seen in 2017

Anyway I don't understand why you think the miners wouldn't like the idea. It's them who will benefit in the end. If the transition to the new protocol is designed well, there will only be a short period where miners won't get coins from the "burnt coins pool". So the proportion of burnt coins could start on a minimal percentage of the fees and then increase over time. Also the delay before the coins are reinserted into circulation by another miner could start with a few blocks, and this delay could become longer each difficulty period until a stable state is reached after a year or so. Once this transitional period has ended, the income for miners will be typically higher than before and they have a predictable minimum income.

When large corporations have a significant stake in BTC, they'll want to support the network's security. Even the average user would likely be willing to contribute something to protect the value of their holdings if block rewards become too small to sustain the ecosystem. This kind of subsidy could be proportional to each person's holdings -- like paying a few dollars a year to keep your $10,000 safe. I don't think most people would object.

I think that's the same idea like @odolvlobo already mentioned about a demurrage. The problem with that may be the competition of altcoins which don't follow that path. Monero for example has already switched to a tail emission, so they don't need this "tax" ...

[mikeywith]

I think you overestimate the miners' power in the game. If the Bitcoin users and holders think that the version of Bitcoin is superior to the version without that burning (e.g. because its security is superior), then there will be enough miners happy to mine the new protocol. And if they don't want to support a softfork you can threaten with a hard fork. The Bitcoin users have always the last word, as seen in 2017

Back in 2017, miners didn't have much of a voice in what happened. It was mainly mining pools pushing for their own interests and the 1–4% of rewards they received. Most of the hashrate came from mid-sized farms located near rivers in China, run by people who were not involved in the debate and likely hadn't even heard of the fork, or spoken English in the first place. That era is long gone. Today, the majority of mining is owned and operated by large U.S. corporations, both at the pool level and in terms of physical infrastructure.

Besides, that event wasn't even remotely about reducing mining rewards, so miners didn't have much at stake. This time it's different -- you're talking about cutting directly into their earnings. Even if you had tried to push such a change in 2017, it would've been much harder than a simple block size adjustment.

Anyway I don't understand why you think the miners wouldn't like the idea. It's them who will benefit in the end.

The theory of "let's pay you less now so you can earn more later" is a bit funny, especially for someone who could earn more now. This isn't just about this specific topic, it's a flawed idea in general. You're assuming that today's miners will have the same market share in the future, which is not true. This approach will mostly benefit miners who have the ability to rapidly expand and outcompete others. As a mid-sized miner myself, I know I won't be able to outpace the large corporations unless Trump bans Bitcoin mining.

 

[Satofan44]

While reading some content, a user reminded me that Bitcoin was supposed to go bankrupt in 2016 for this very reason. It didn't happen, and here we are again with the same concerns by different people who probably do not remember this story (I am at fault as well).

This doom scenario was supposed to happen after the 2016 halvening when Bitcoin miners had not been not profitable for almost 2 years from 2014. Lots of mining companies went bankrupt in those years. There were tons of articles about how after the 2016 halvening, BTC was doomed.

   By mid-2014, the high revenues of 2012 and 2013 are countered by high expenses, leading to a negative net cash flow from that moment on.

https://link.springer.com/article/10.1007/s12525-018-0308-3

Bitcoin developer Mike Hearn on January 2016:

   The death spiral begins

    Maxwell and the developers he had hired refused to contemplate any increase in the limit whatsoever

    clock was ticking

https://blog.plan99.net/the-resolution-of-the-bitcoin-experiment-dabb30201f7

What happens is mining costs converges to the price of electricity and/or competition wipes out inefficient miners who don't innovate, find cheaper energy and locations.

Since 2015, the miner rewards have been cut by -87.5% but BTC price has gone up by 47,100% so miners rewards are more than enough. Fees alone will be enough reward the miners when block subsidies end. You can also slowly increase the block size over long time frames if needed to increase the subsidies if needed.

Source. https://old.reddit.com/r/CryptoCurrency/comments/1m15ao8/what_happens_when_all_21_million_bitcoins_are/n3fvd42/

Shitcoiners and corrupt assets like Hearn already tried to destroy Bitcoin's model and implement infinite inflation before.


Still does not look like a big deal to me. Volatility in rewards in an asset that is still volatile seems like expected behavior. It would be an issue if there was no difficulty adjustment mechanism. 

[MeGold666]

Sometimes it's better to sell an old car and buy a new one that will drive you from point A to point B more efficiently, with all the previous generation's problems fixed.  

[Wind_FURY]

While reading some content, a user reminded me that Bitcoin was supposed to go bankrupt in 2016 for this very reason. It didn't happen, and here we are again with the same concerns by different people who probably do not remember this story (I am at fault as well).

This doom scenario was supposed to happen after the 2016 halvening when Bitcoin miners had not been not profitable for almost 2 years from 2014. Lots of mining companies went bankrupt in those years. There were tons of articles about how after the 2016 halvening, BTC was doomed.

   By mid-2014, the high revenues of 2012 and 2013 are countered by high expenses, leading to a negative net cash flow from that moment on.

https://link.springer.com/article/10.1007/s12525-018-0308-3

Bitcoin developer Mike Hearn on January 2016:

   The death spiral begins

    Maxwell and the developers he had hired refused to contemplate any increase in the limit whatsoever

    clock was ticking

https://blog.plan99.net/the-resolution-of-the-bitcoin-experiment-dabb30201f7

What happens is mining costs converges to the price of electricity and/or competition wipes out inefficient miners who don't innovate, find cheaper energy and locations.

Since 2015, the miner rewards have been cut by -87.5% but BTC price has gone up by 47,100% so miners rewards are more than enough. Fees alone will be enough reward the miners when block subsidies end. You can also slowly increase the block size over long time frames if needed to increase the subsidies if needed.

Source. https://old.reddit.com/r/CryptoCurrency/comments/1m15ao8/what_happens_when_all_21_million_bitcoins_are/n3fvd42/


Shitcoiners and corrupt assets like Hearn already tried to destroy Bitcoin's model and implement infinite inflation before.

Personally I believe that "infinite inflation" is GENERALLY not the problem if percentage of inflation per year vs. total supply is in a manageable level to prevent too much dilution. Transparency and a consistent emission of a currency as agreed upon through community consensus is definitely a property of Sound Money.

Still does not look like a big deal to me. Volatility in rewards in an asset that is still volatile seems like expected behavior. It would be an issue if there was no difficulty adjustment mechanism.  

Probably not, but debating and discussing for a sort of tail-emission for Bitcoin is good "in case" the network will need it. It's better if it doesn't, but it's also good to be ready.