Under proof of work you essentially end up "staking" money by buying ASICs/GPUs that take 9-12 months to break even. In general, miners are going to spend almost as much as the block reward, whether that's in hardware and electricity or in opportunity cost. http://www.truthcoin.info/blog/pow-cheapest/
And many people are "hodling" cryptocurrency as a long-term store of value anyway so presumably they aren't concerned with the velocity of money.
Long term hodlers who haven't locked their coins into staking can respond quickly to network events such as Bitcoin-cash. People who are running mining businesses can at least sell the business if they suddenly need access to a lot of capital. With proof of stake, it really is locked up for whatever period of time.
If your staking period is just a couple of weeks, this is probably less significant than owning a mining farm. But if your staking period is several months, the opportunity cost probably is greater than mining or hodling, because you sacrifice a lot in terms of flexibility.
When you put it that way, I expect to see a secondary market for stake. This could be implemented by selling the private key or by delegation similar to a mining pool.
Selling the private key would be a bad idea since you can't guarantee the seller destroys it. But you could have the staker address be a contract, with an ownership that can be transfered.
You could use an HSM to hold the private key and then sell the HSM. Overall it's still a bad idea to buy an HSM because you don't know what other transactions have been signed, but maybe the HSM keeps a history of everything it has signed as well.
Which, if you can still sell your staked coins, what is the point of locking them up in the first place?
But you are essentially staking "off chain" by buying the gear in fiat money. So there is a built-in exchange rate arbitrage. And the on-chain velocity of value is not slowed down. This isn't true in ETH POS.
Ethereum's block hash algorithm is more memory intensive than bitcoin's and benefits less from ASIC work -- please correct me if I'm wrong but I don't think that it faces the same problems.
The same thing applies, you have to buy/power GPUs and decide they are best utilised mining ethereum, mining another coin or powering your new startup.
You're right and it's proven pretty robust at that. This is a good proof of work, because it means that anyone with commodity hardware can set up a miner, and the playing field is pretty level.
Assuming everyone has existing funds to acquire mining hardware.
Large existing expendable cash has a greater advantage to mine and mint. Mining pool operators are also in a position to manipulate minting and transactions in the network.
And many people are "hodling" cryptocurrency as a long-term store of value anyway so presumably they aren't concerned with the velocity of money.