A limited blocksize won't result in fees climbing higher in the long term. It would result in a short term peak followed by abandonment and loss of value for be

A limited blocksize won't result in fees climbing higher in the long term. It would result in a short term peak followed by abandonment and loss of value for being unusable. People pay fees for that which they need and find more desirable than the alternatives - an effectively unusable Bitcoin won't be worth paying fees for. And assuming unlimited block limits will lead to no fees paid is ridiculous and senseless. Why would miners willingly use up more and more bandwidth as the subsidy keeps falling through accepting more and more fee-less transactions? They'd be losing money! First of all, the miners should be requiring that the total level of fees reaches to a certain minimum level which exceeds what a plausible coalition of adversaries might try to gather, so that the mining efforts they can pay for will exceed that level. Then you require that the minimum fee per transaction is that minimum divided by the largest number of transactions that the network reasonably can handle per block. Then miners prioritize transactions with larger fees and publishes estimates for how fast people can anticipate transactions to be processed based on fee density, in order to incentivize keeping the total transactions fees to a reasonable level. Fee densities so low that they given the network capacity makes it easier for attackers would then be discouraged.