I don't think the people paying for security have to be the same as the people benefitting from it.... But backing up, thank you for writing all that. I think I understand and have had similar worries.
Actually I think the hodlers, who aren't the active users, are paying for security indirectly. By hodling, they constrict supply, which keeps value high, which keeps the miners happy with what they get paid. If you push the scenario to the opposite extreme, where nobody keeps savings long term and spends their while income as fast as they get it, then you would effectively have in increase in supply of btc, which would bring the value of each sat drastically down. Then you would be in a similar miner death spiral, just from the other direction. This would be the same effect that the "velocity of money " currently has in economic calculations - that is, the bank reserve ratio is the inverse of the multiplier of the absolute number of dollars loaned - so if the reserve ratio is 10%, and a bank makes a loan of $1 million, then the money supply effrctively increases by $10 million. Therefore, not spending btc assists in retaining its value, which is good for mining.
I think the solution is similar to this - just as constricting money supply keeps value up, also constricting block size keeps fees up. If, after the coinbase drops to zero, the miners can't stay in business and it appears that hash rate will fall below a secure threshold, then the fix is to decrease block size. IMO it should never have been increased to begin with.
Does that make sense? Maybe there's something I'm still not seeing. An aside, but I think there's a point where increasing block size becomes genuinely dangerous, if it causes miners to stop mining honestly and instead attempt a reorg. You can always make blocks smaller without danger, but not bigger. I hope we don't see any more of that.
Replies (3)
Well, they're not paying anything; they're not incurringna cost to cover the security they derive.
Take this idea to an extreme. If *everyone* holds, then the value goes to infinity, right? Well no, because miners stop mining entirely because they won't get paid. There's a curve here, it's not a linear relationship. It's not long term viable, pressing the price up only generates more miner revenue for the same number of coins to a point, and there's a point at which that upward price pressure is outpaced by reduced number of transactions. This is like a cold staking scheme and is not stable or long term viable at all.
Ultimately, the only way to pay for the ongoing cost of security is to have an ongoing cost to holding.
As far as block size... The goal is to secure the network by creating blocks with high difficulty, as well as to process transactions. The smaller the block size, the less transactions you can process. This accelerates what I talked about above since transaction fees double if the block size is half, and as I explain above, constricting supply isn't enough to make sure miners are paid well enough forever. Take this to the extreme too: a block size of 0, in a supply capped coin, will ensure that all miners stop mining forever because there will be no transaction fees. Unstable, not long term viable, similar math and incentives as above.
I don't understand how a bigger block size incentivizes miners to attempt a reorg. If you can explain to me how that would work I'd like that, maybe I'm missing something but I don't see it.
Increasing block size could incentivize a reorg because fees per block fall, so some number of miners become unprofitable - the most expensive/least efficient miners. They're still hitting nonces at the same rate as before, butwity the payout lower, it could make sense to start turning off the miners with the highest cost per hash. At some point, you could have a lot of hardware sitting around, and at some point its enough to attempt a reorg. I'm not super familiar with this stuff, so I'm just conceptualizing. Idk for sure but I think the bigger the reorg (further back in time), the more hash is required. So if hash is coming offline because fees are low, then there's a curve you could plot where hash profitability is negatively correlated with the depth of a potential reorg. So expanding the block size has to be done very carefully, at a slower pace than new hashrate coming online.
The opposite would be no problem - reducing block size increases profits per hash, so previously unprofitable miners could be profitable again. Of course there are limits, like if the economy is dependent on opening and closing lightning channels and block size constricts the possible throughput regardless of fees.
As far as block subsidy, it will likely maintain the value it has now indefinitely. Most people seem to think that the Satoshi will be the smallest denomination forever. It won't. As a matter of fact if a penny is worth anything by the time Bitcoin hits 1 million and 1 dollars a coin, we have a denominational issue. So extrapolating from there the more the value climbs from both inelastic supply and rising demand, the more decimals become commonplace. The blocksize now is pretty much perfectly designed to handle 5000 countries worth of settlement transactions per 10 minutes.
As the incentives of bitcoin kick in at the national level, states that hodle first become more powerful than the nations that encompass them. Making the cultural separations more important and further dividing the current sub 200 countries into more. This also decentralizes mining and distributes block fees among smaller parties. The cost of a large mining op just won't be able to compete with hundred of thousands to millions of individuals running 10TH/s systems at 15-17J/TH.
As far as economic velocity, that never gets handled by an asset layer. That is the final settlement for when the trust of economic actors falls or conflict arises. It is utterly staggering how many billions of transactions happen on the back of "trust me, I'm good for it." With layers like lightning it's more like "Trust the code that you can read for yourself, I'm not going to force close the channel while you're offline with a previous channel state (If I get caught I'd lose all my sats)." Much more complicated but, much less trust is necessary. The restricted block size simply makes securing the network way more cost effective. What do you think is easier auditing every ounce of gold in the world or validating the bitcoin timechain? Now THAT'S security.