Playing devil's advocate, for a moment (2nd time today, I just notice)
Not challenging the 'minimum valuation required to perform as medium of exchange' argument. But why can't the ledger be separated from the currency function? For a number of imaginable functions in the future, a minimal share of the supply would be enough to perform the ledger function (e.g. contracts).
In order not to gloss over this: there's still the valuation through required security. If contracts on blockchain have a certain total value, the network needs to be at least protected to make an attack economically unfeasible. Since miners are economically motivated as well, this will provide a security-based minimum valuation of Bitcoin in the process.
Still, the two values need not be the same, so the question remains: why can't the 'ledger' function not be separated from the 'currency/money' function?
You could separate the currency from the ledger, but to what end? Seems to me like that would be making the system and incentives needlessly complicated and would still require you to purchase tokens in order to pay your way into the ledger. In some ways, that is exactly what Ethereum has done. Either way the tokens are still required to make the ledger function, so why make them separate? Perhaps someone has an idea as to why this would be desirable?
My mistake... could have phrased that more clearly, I guess:
I don't suggest we "split up" Bitcoin. My question was, "what prevents the rest of the world from adopting Bitcoin, the distributed ledger, but (mostly) ignoring Bitcoin, the currency"?
The usual argument goes "the two functions cannot be separated, so the above is impossible". I'm asking, why? To me it looks like a possibility, and the implications for total valuation in the long term would be drastic.
Because there would be no incentives for the miner to secure and maintain that ledger?
Maybe I have your argument wrong but this is how I understand it
Again, apologies. Maybe not phrased clearly enough.
The side remark in my original post ("In order not to gloss over this") was meant to address exactly this point: even if the world only makes use of the ledger function of Bitcoin (and as such, no significant valuation floor is provided by that use), keeping the "ledger only" network secured will provide such a floor.
However, valuation in that world could be substantially lower than in the one that includes usage cases medium of exchange / store of value.
a) Valuation from the above two usage cases is missing.
b) Effort needed to provide security of the network could well be not in a linear relation to total value of ledger usage:
- double spends are less of a concern if contracts are to be settled (confirmation time less of a problem)
- there is more than just a monetary requirement to overcome to successfully mount an attack on a network the scale of Bitcoin (hardware procurement, mounting the attack without raising suspicion)
The point I'm getting at is that the total value of contracts settled through the ledger can be higher than the total cost to secure the network, as long as the cost to mount a successful attack on the network is prohibitively high for any individual actor.
So the total valuation / market price per unit could well be below the current (optimistic) estimates that include the medium of exchange / store of value usage cases, even though the security requirements for the ledger usage alone provides
some valuation.