Here is a different scenario - how would you deal with this:
Service loses money (hack, lost keys, whatever). No worries (for the customers), there is escrow. Service says "we fucked up, escrow please refund our customers". Escrow runs away with the funds and doesn't refund. Whom do you flag and why? Service? Escrow? Both?
That would be seriously fucked up. In this scenario, the escrow deserves a type 3 Flag for sure, to be created by the service. After that, I'd say the service should still refund the customers, or they can create a type 3 Flag against the service. Unless the customers agreed to only rely on the escrow before they paid the service, but that wouldn't make sense.
So it sounds as if we would be flagging the service in this hypothetical scenario even though it didn't scam (in theory; assuming there is no collusion, lying, etc), and some argue that in the WWM scenario the service should not be flagged even though they (presumably) scammed or at least got hit by a bus and had no communication contingency.
There needs to be consistency in how we interpret these implied contracts. I think in both cases the service that took the money and didn't pay back needs to be flagged with one of the contract flags, regardless of what any escrow may or may not do. If you steal something it's a crime regardless if the victim gets an insurance payout. I don't think a new type of flag or even a change in wording is needed, but I'm not a lawyer, just occasionally pretend to be one.