The usual explanation for how proof-of-stake works — compared to proof-of-work — is that it uses the scarce resource that is the chain’s own token, as opposed t

The usual explanation for how proof-of-stake works — compared to proof-of-work — is that it uses the scarce resource that is the chain’s own token, as opposed to the scarce resource that is electricity. The fundamental problem with this approach is that, in the absence of consensus, a token is not scarce at all — but rather its supply is unlimited — since infinitely many tokens exist on an infinite number of valid chains. So, in order for a token to be limited in supply, consensus on which chain is the canonical one must already exist. Thus, proof-of-stake reaching consensus depends on consensus existing beforehand. Scarcity of a digital decentralized token requires consensus on which chain to view as the truth. Therefore scarcity cannot be used as a requirement for reaching consensus.