Something occured to me today.
BTC's idea is that eventually after many halvings the fees must replace block rewards.
1) We know that BTC's transaction cap is somewhere around 400 000 transactions per day
2) We know that the current level of security provided by the miners is paid for by around US$17m daily (see "Reward last 24h")
3) $17m / 400 000 = $42 -- that's the fee that on average each transaction must pay to have the current level of security that Bitcoin (BTC) has when block rewards go to zero
4) If we zoom in to December 2017 for median transaction value, we'll see that it was about $1000-5000 - that's the typical amount people moved when typical fees were at $15-30
5) There are only 2.5 million BTC addresses that have $1000+ (bitinfocharts, see "Addresses richer than 1/100/1,000/10,000 USD") ($1000 being the lower bound of median transaction value in December 2017, when fees were in fact around $15-30)
6) 2 million people at 400 000 transactions per day means each one only allowed to make 1 transaction per 5 days.
So, here the endgame for BTC:
Bitcoin (BTC) is a network with capacity for about 2 million people (approx. the population of Hawaii), each waiting at least 5 days to make one single transaction of about $1000 in value with a commission of at least $40 (which is more expensive and slower that a typical international bank transfer).
That's also the uncomfortable truth for the Lightning Network: at BTC's endgame, opening a channel would cost $40 and closing a channel would cost $40 more.
[link] [comments]
source https://www.reddit.com/r/btc/comments/eska86/btc_fees_endgame_40/
No comments:
Post a Comment