There are a few answers to the long-term security budget:
1) We could maintain fees through compression, that is, by making each on-chain transaction representative of many off-chain transactions. This is beginning to happen with the lightning network, and other technologies are in the works to advance this concept.
2) Mining will become more integrated with other industrial and residential processes, which will reduce the cost of security, from a compensation standpoint. Think bitcoin miner water and space heaters in many homes, etc.
3) It's also likely that in the future institutions that rely heavily on bitcoin will voluntarily subsidize security to some extent, for the same reason they invest in vaults for other instruments. Personally I'm planning to start running a lottery miner at home just for the fun of it.
(2) is interesting, but I'm not sure if resistance heaters will remain common in the long term, as heat pumps become more widely available?
Also this seems like a sort of free-electricity scenario (mining with electricity that was going to be used anyway), but I'm not sure free electricity would reduce mining costs? Wouldn't hardness adjust so that it then becomes all about hardware CapEx (assuming amortized CapEx remains significant)?
I figure mining costs are around 1% of the market cap, and can't go that much lower without serious security risks. So the ecosystem needs to fund that in some way, whether it's through inflation, fees, or donations.
With fees, I also worry they'll be too unpredictable. Like there might be enough demand to justify $20B/year in fees, but it's a function of supply as well, and if scaling solutions like Lightning Network work "too well" (with people rarely needing to settle on-chain) we might end up with way lower fees.
1) We could maintain fees through compression, that is, by making each on-chain transaction representative of many off-chain transactions. This is beginning to happen with the lightning network, and other technologies are in the works to advance this concept.
2) Mining will become more integrated with other industrial and residential processes, which will reduce the cost of security, from a compensation standpoint. Think bitcoin miner water and space heaters in many homes, etc.
3) It's also likely that in the future institutions that rely heavily on bitcoin will voluntarily subsidize security to some extent, for the same reason they invest in vaults for other instruments. Personally I'm planning to start running a lottery miner at home just for the fun of it.