If the banks hold subordinated bonds issued by NAMA, in theory they will feel some pain if NAMA is in the red when wound up.
Remember how it is claimed that this going to work. Lets say they go for the 5% ratio. NAMA will give the banks roughly 52 billion of floating rate bonds (there is still no clarity on the structure of these bonds) and 2.5 billion of subordinated bonds (presumably fixed rate) in exchange for the toxic loans.
Each year, NAMA will have to pay the banks coupons (say 750 million initially on the floaters and 200 million on the subordinated bonds).
At the end of the ten years, NAMA will have magically converted the toxic loans into a war chest of cash - over 55 billion according to the "business plan". It will then hand over this cash pile to the banks to honor the bonds it has issued.
If - surprise, surprise - NAMA somehow fails to be able to sweat the equivalent of the combined net worth of Bill Gates and Warren Buffet out of toxic loans to Irish builders in the 10 years (all the while paying the banks 1 billion a year in coupons which is likely to double that if interest rates rise even 2% or so), then NAMA won't have enough to pay off the bonds (i.e. the banks).
However, any shortfall under 52.5 billion will be made up by the exchequer. So the banks are guaranteed to get 52.5 billion. Were NAMA not to have amassed the full amount (55 billion) then the banks would stand to not get paid the last 2.5 billion (of subordinated debt). This is chicken feed in terms of "risk sharing".
Of course, deferring the pain suits most of the main players involved in this mess at the moment particularly politicians, bankers and developers. Ten years is a long time in politics and it's only then that the bill for NAMA (likely to be 10 or 20 billion, I suspect) will be presented. In the mean time, the banks will have been paid between 10 and 20 billion worth of coupons, 2.5 billion will have gone on accountants, solicitors and consultants and there will be endless opportunity for political meddling (social housing, etc.).
The bill will be paid by tax payers. The final sick joke is that it is unlikely that NAMA will improve credit conditions in the country.
What is interesting is that in admitting that subordinated NAMA bonds would have to pay junk-bond like coupons, the minister is telling us what he really thinks of NAMA's chances of success.