In a Proof-of-Stake system similar to bitcoin, a large number of coins could lie dormant and accrue 'coin days'. If these coins are in escrow, like on Mt.Gox, their 'coin days' could be used to do an attack on the network.
Thus any large escrow service would be a threat to the network, in addition to large miners.
Thus either escrow services must pay interest, or the need for escrow should be eliminated by a better block chain design and p2p exchanges.
It's true that an exchange or wallet service could use it's wallet to launch attack on proof-of-stake, although unlikely. The current plan is to implement reorg depth limit and relegate checkpoint to be advisory be default, so if this type of attack (considered to be equivalence of 51% attack on proof-of-work) occurrs users can subscribe to checkpoint so that transaction processing can continue on block chain.
Miners don't have a play in double-spending attack, unless they wait and become stake owner. Security comes from proof-of-stake, proof-of-work only provides minting. Please don't confuse ppcoin's design with other proof-of-stake proposals. Our design is the only one that gives full respect to the concept of proof-of-stake and is the only one that actually has an implementation rather than just talks.