[jpsalm]

Can someone explain this to me?

The year is 2150 or whatever -- bitcoin is fully mined. There is a metric asston of mining hardware out there that needs to be running to secure the chain -- presumably all coming from transaction fees.

Not only to transaction fees have to scale to cover the entirety of block rewards but they must continue to grow over time to compete with increased hardware efficiency in hashing.

Does it not follow that:

a) transaction fees rise to recoup not only current cost of mining hardware, but R&D and deployment of new hardware over time to keep the chain secure. As people realize this they convert their bitcoin to something else to avoid bag holding.

OR

b) transaction fees cannot support the mining hardware as is, miners sell of their hardware to recoup their costs and the chain becomes vulnerable.

I confess I might be missing something but I don't see how a fully mined coin is stable.

[X6S1x6Okd1st]

You are right, the security of the chain should always be proportional to fees paid.

The security of any given transaction is a little different. In practice we haven't seen that many double spends against chains that are defacto vulnerable (e.g. any chain that uses the same PoW as bitcoin is always vulnerable to an attack).

Assuming that PoS goes well for ETH it might make sense to start moving away from PoW coins before the fees drop to worrying levels.

[rcxdude]

It seems quite unlikely for bitcoin to successfully change anything about the protocol: even simple changes have been failures.

[rodiger]

As of right now yes, but when there are more stakeholders and clear/imminent threats to the chain there will likely be more discussion. All proposed forks thus far have been unnecessary for the short-term success of the chain.

[rcxdude]

Yeah, I think this is a strong reason why bitcoin as a 'store of value' makes little sense over long time periods. If ultimately you need to burn a substantial fraction of the value of the pot to keep someone from running away with it, in the limit where no more bitcoin is being created then you basically have an asset which has large maintenance costs but no productivity. If a cryptocurrency is being used to transact reasonably efficiently (which bitcoin absolutely is not), then there's value being created by this which makes the costs worthwhile. But if it's not and the idea is people mostly sit on it, it's hard to see how it works out in the long term. (I mean, gold is the same to some extent, it costs some money to run a vault, but what's the relative costs? certainly it doesn't cost as much to keep the entire world's supply of gold secure as it does to run the bitcoin network, and gold's market cap is 20x that of bitcoin)

Unfortunately I think there's a lot of supply of coin and cheap electricity left to keep fueling it for far longer than it makes sense.

[rodiger]

A bit late to this but a few things:

If BTC is not dead by then, it will likely have some sort of active user base, meaning non-zero fees.

If the fees are low, verifiers will stop verifying.

If enough stop verifying, the prize pool grows (more network congestion leads to higher fees). This eventually stabilizes at some equilibrium (likely leading to a weaker chain, if profit is the only motivator).

It's also worth noting that if BTC has seen enough adoption to be sustainable for the next 100 years, there are likely many many stakeholders who could/would mine for "altruistic" reasons. E.g. financial institutions and other large holders have incentives to keep the network stable.

Hell, maybe by then they get enough people on board for a BTC 2 that's some weird Ethereum competitor. It's really hard to theorize on that long of a time horizon.

[jpsalm]

The cycle is self-reinforcing though.

As transaction fees increase to cover the cost of security (even an "altruistic" (self-interested) financial system needs to pay for energy) users are less likely to transact which is positive feedback on the transaction fee.

If deflation cannot outpace this then Bitcoin effectively dies. The longer you leave funds on the chain the more you will pay to pull them out. If deflation does outpace security cost you have runaway infinite deflation that is no longer tied to a from of scarcity but instead the cost of securing the chain. That doesn't seem great either.

> Hell, maybe by then they get enough people on board for a BTC 2 that's some weird Ethereum competitor. It's really hard to theorize on that long of a time horizon.

Possible, but I don't know if I'm optimistic about humanity's ability to generate collective consensus in the face of crisis.