How to pay for ECB failure?

Buiter likes to provoke and think the unthinkable. He does do that exactly in his CEPR piece that ask the question whether ECB can be in need of bail out after taking dubious assets as collateral to its liquidity-enhancing loans. He argues that EU27 governments should come up with the fiscal formula splitting the burden of ECB bailout.

He has a point. However, what seems to be much more likely than ECB going under is one of the medium sized banking groups that control banks in EU10 going belly up. Many of them are big enough in quite a few EU countries to be counted as too-big to fail. So now, you have a bank operation, say in 10, of the EU27 countries in need of bail out. Who is going to pay for it? The home-country government? A syndicate of home- and host-country government? How would they share the bill?
Time is a precious commodity, when a bank is sinking. There might be a window of a few days to agree on the cost-sharing formula, but it is unlikely that governments could strike reasonable agreement under such time pressure. Without pre-agreed formula, the ad-hoc agreement would be a sure recipe for endless disputes, litigations and arbitrations, that would leave noone but bunch of well paid lawyers happy. The political fallout of such disputes on the European integration project would be ugly (Euroskeptics might be the only ones to rejoice).
An alternative would be to let the ECB to step in. Instead of bunch of squabbling governments, the ECB could lead the rescue efforts in multiple EU countries. Given the nature of the financial integration within the EU, shifting the task to ECB would increase chances of a successful solution. However, the ECB would need to pass the costs to governments. At the end, this might be the more important reason to start thinking about the formula for splitting the costs of bail out.