4 ms·
We only infer value through market transactions, and we are actually inferring a bound on the surplus rather than the precise value being exchanged. It does ma
by alphazard 2y ago
We only infer value through market transactions, and we are actually inferring a bound on the surplus rather than the precise value being exchanged. It does make sense to draw distinctions between different kinds of value, and often "intrinsic" and "extrinsic" are used, even if they may not be great terms. I can give a few examples.
- In options pricing. The intrinsic value is determined by what the contract says happens at expiration. The contract is enforced by the authorities like any other legal contract. The extrinsic value is the name we give to the value that must exist to make up the difference between the intrinsic value and the current market price. It's sort of like the "dark energy" of the option.
- Gold is said to have intrinsic value because it can be used as a material in ways that other material's can't. It can be used as an input to whatever industrial process regardless of what humans have to say about it. There is additional value implied by the price of Gold, just as there is with the price of an Option.
Intrinsic value means something like "if the other market participants disappeared, the value the last person holding the thing could still get from it". Maybe as a matter of semantics, you would exclude this from the "economic perspective". In the case of Bitcoin, intrinsic value may just not exist (as you seem to imply).
- npoc 2y agoIntrinsic value for an option's strike price is a special case, as the value is fixed to a historical moment by prior agreement. Correct me if I'm wrong, but it seems from the rest of your post that you see intrinsic value as "non-economic usefulness". But gold's primary usefulness is as the best physical hard money due to it's fundamental properties: scarce, divisible, indestructible, recognisable, fungible, assayable (its weight, colour and softness) However, your definition states: > "if the other market participants disappeared, the value the last person holding the thing could still get from it" and money has no use/value if there is no one to transact with (i.e. no other market participants) So if we were to agree that intrinsic value is the usefulness of something but with the exception of any purely economic/market use, then to use "it has no intrinsic value" as an argument for why something would not be a good money (as the author does) makes no sense, as it ignores the exact usefulness that would make it a good money.