Lots of things at play here. I'm by no means an expert, but here's my thoughts:
I have only heard of it being handled as a reverse auction, which is the only way it makes any sense. A bank sets the price they feel the MBS are worth, and the treasury starts buying at the lowest levels moving up with their $700 billion checkbook. This encourages a bank in need to price low enough to get bought. It also stops, or at least limits a banks willingness to overstate values. Think of it this way, if you had a giant pile of garbage weighing down your company, and the ability to throw said garbage overboard, you aren't going to risk it not getting bought because you want to overstate the value. This is your get out of jail free card if you're a CEO. You don't throw it out with the bath water.
As for banks specifically knowing the true 'value'. Of course they don't. They are valued off of past models, and current performance, but if you've got a vintage of early 06 95-100% ARMs, who knows how they are going to react come adjustment time. Same thing with Option ARMs that aren't into recast. Someone could still basically be 'renting' a house that is upside down 30%, and when the recast comes, they walk if thats the case. Thing is, recast may not be for another 2,3, or 5 years. How can you properly value that? In order to value, you need a basis of performance. Its the same case as subprime MBS "performing" in 03, 04, 05 etc because when someone got in trouble they could simply refi. So while there was a bundle of pre-payment in the tranches (hence, the requirements of PPP, to safeguard investor returns), there weren't defaults. At least not to catastrophic levels, until the value well ran dry and people couldn't simply churn their loan when they got in trouble. You can't value loans that haven't hit the breaking point, because you are valuing them on a false pretense.
For me, if I'm a CEO, I get with my CFO, my accountants, and my board and say look, we've got X in MBS, port loans, CDO's etc that is either A.) Exotic, B.) ARMs yet to adjust, or C.) Lower credit quality than something we would approve today. On any of those packages I'm finding my pain point, to price as cheap as possible to get it off my books and move on. I don't think this is a perfect situation, but I think it's the smartest thing we've done so far to potentially fix the problem.
AC