One of the mantras I hear repeatedly in this community is "not your keys not your coins", outlining the importance of self custody. In this regard, is it actually realistic for majority of people to self custody their bitcoin on a blockchain that only supports ~ 7 * 60 * 60 * 24 * 365 ~= 221M tx/year? Doesn't this mean if 1B people wanted to hold their own funds it would require at least 4 years? - this is not even including other miscellaneous transactions.

Lightning seems like a promising layered solution, and I very much agree with this design. But, my understanding is that one of the fail safes to bad actors on the lightning network requires you to open and close channels each consuming a transaction on the main blockchain. While transactions within the lightning network are designed to be infinitely scalable, it doesn't seem to be the case when you need to regularly write to main blockchain. Is it feasible/secure for someone to custody bitcoin only on the lightning network?

Given these parameters, how realistic do you think it is for bitcoin to really become a global money? Did I make any incorrect assumptions? Please enlighten.

  • Your transaction per year data seems to be made out of thin air, why not instead look at something practical. Try to take your own coins into self custody, if it is outrageously difficult then maybe there is a problem. I know a lot of people who have taken self custody in the matter of minutes for less than a few cents in fees. So I really don't even understand what the concern is when you add up the lack of evidence presented here.
    – Poseidon
    Commented Jan 21, 2023 at 16:51
  • Lightning is not a perfect solution, I ran a node for about a year and because I did not know enough technical skills at the time I nearly got rekt. I had to contact developers to help me negotiate a channel situation from fund loss. The point is that lightning is still in its early stages and the only reason its talked about is because it has some promising characteristics, it is nowhere near feature complete. There are other popular layers: liquid & rootstock. These both have faster transactions and other cool features. There are more layers being developed, security takes a LONG time
    – Poseidon
    Commented Jan 21, 2023 at 16:55
  • I dont know how you arrived at your formula to calculate tx per year but in fact one transaction can be made or signed off on by several people at once, as well as the addresses in those transactions can be controlled by multiple people. It is immensely scalable if you really do some research into it.
    – Poseidon
    Commented Jan 21, 2023 at 17:05
  • @Poseidon BTC transactions per second is reported by various sources to be between 3 and 7, you can search this. That's the first number in OP's product above. You can also find charts of transactions per day over several years from several sources, it's a few hundred thousand per day on average, a little less than (7 * 60 * 60 * 24) above.
    – tsj
    Commented Oct 12, 2023 at 19:44
  • Sure but those transactions can be multi-signature contracts in which case it looks like one transaction while it is actually scaling to facilitate n number of off-chain transactions at the same time.
    – Poseidon
    Commented Oct 12, 2023 at 20:52

2 Answers 2


You are right. We would not be able to have every person on Earth start using Bitcoin from today to tomorrow. But “necessity is the mother of ingenuity”. Just as it would have been impossible for millions to stream movies to their homes via the internet at the end of the last millennium and we have found ways to facilitate that by now, Bitcoin is organically adapting to its users’ needs.

  • Blockspace demand was hitting the limits, so businesses started batching their payments and the segwit softfork introduced more blockspace efficient output types which also simplified implementation of the Lightning Network as a payment layer.
  • Lightning scales payments but not users, so there are now e.g. two proposals how to build custodial Chaumian E-Cash systems on Lightning, and one how to build a distributed currency exchange layer.

Building a protocol is a 20 year project—I’m sure we will bump into a few more challenges and find solutions for them along that way.


If the number of users wanting on-chain funds increased dramatically, on-chain cost (those fees) would skyrocket, causing some of those users to desist wanting to be on-chain, maybe expelling some others to cheaper alternative chains.

For most people, custodial funds on L2 would be good enough. If their available funds increase, at some point (of increase pace) they'll be able to offload a part of them funds down to the chain for sovereignty and security.

If the funds are not high enough but the user persists on getting them down to the chain, that's some kind of educational or geeky use case. They can use testnet for the earlier or litecoin for the latter.

Naturally this topic is affected by price, but one can reasonably expect the software would be forever adapted to fit on-chain "small enough" value transactions such as maybe real estate changing hands level transaction. If not, eventually alternative chains might start leeching economic value out of Bitcoin, Bitcoin becoming a niche chain for big entities only, more dangerously eventually eroding maintainer input to the point of risking a catastrophic collapse due to brains work having migrated to alternative, more human size friendly chains.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.