PoS does not reinforce historical consensus. Every subsequent block in a PoW chain makes the history below it more secure because the cost of reversing it is superlinear in the number of blocks built on top. In PoS, this is not the case, the cost of producing a block is a constant, therefore the cost of reversing history is a constant
But is it really so?
For the record, I'm not technically familiar with either PoW or PoS systems but let's assume that there is only one miner left in Bitcoin and he waits long enough till the difficulty drops. Won't he be able to do anything with the blockchain including reversing transactions? If he will, how does every subsequent block make the whole blockchain more secure then? As I understand it, he will be able to completely undo all transactions. What am I missing, in layman terms?
So when the rest of the network come back up they can just as easily disregard this fake chain (by the single Miner) and carry on with their own chain and be the main chain because the protocol recognizes the chain with the most work put in as the valid chain. And no matter how small/large the hashpower of the miner network is, even if they all collude they won't be able to touch funds from a public address whose private key they don't know. The only thing they can do is a double spend attack if they can consistently write their own blocks to the blockchain.
Also if only one miner is mining the coin it's already dead anyway, so the lone miner can do whatever they want and no one would care.