In most changes the "big picture" is good enough, but the devil is in the detail; that is where the problems come to light. It seems to me that in this case of breaking Greece out of the Euro there are serious problems in the "big picture", ones that will only compound in the detail.

The big question is how one draws a line around the assets and liabilities that will comprise the New Drachma?

When we created the Euro it was easy. ANY & EVERY asset and liability in each of the component currencies were converted. In this instance, just SOME of the Euro will be broken out. The key question is what and where?

Simple domestic Greek assets and liabilities are not the problem. It is the more international aspects that create the challenge. Imagine a Greek company that legitimately and reasonably took out a euro loan with a Belgian bank (I will avoid using Germany as the example) and then used it across its European operations, with the balance held in a head office account with a Greek bank in Athens.

It would seem that the portion in Athens is a no brainer, but what about assets in France that were funded by the loan? Similarly can one force the lending bank in Belgium to redenominate its loan to New Drachma, with all that could imply?

Unless we corral all the relevant assets and liabilities we will have an unbalanced set of accounts. So what do we do with the mismatch?

The only answer I have only come up with is not simple and could have some serious implications. I don't think that the EC will be able to issue rules that inherently create matched assets and liabilities in the New Drachma. This means that in keeping with the principles of double entry accounting there will need to be a sweep, a collection and consolidation of all the mismatched balances. This, I expect will have to be cumulative ie local banks reporting to their own central bank who then, I expect will need to work with the ECB. I expect that the ECB will have to absorb the full impact, taking the resultant exposure and providing corresponding compensation to the rest of the banking system.

Without this there could be considerable exposure in each regional (eg UK, US, etc) banking system. This would be unacceptable and create even more uncertainty and instability with each region trying to look after their own interests and exposed to the actions of speculators and hedge funds. This is not a recipe for financial stability.

Where any rules leave doubt or ambiguity I would expect enactors to take what they consider to be the most advantageous choice. I would expect the New Drachma to be pretty weak and that non-Greek banks will try and keep as many of their assets in Euros while converting their liabilities into New Drachmas.

Whatever happens it seems that identifying Greek domicile will be critical. For the banks this will hinge upon the static data held for clients, instruments and accounts. Now we know how notoriously inconsistent that data is. We spent months cleansing the data before creating the Euro, but we will not have the same run up here.

I know that a number of Banks are already in data clean up mode for KYC (know your customer) requirements, but expect it to take years. I am also aware of many debates around identifying the country of issue for a number of instruments , especially for larger supranational organisations.

Now while the first point still means I don't think we know the boundaries, I think it is clear that there needs to be an urgent focus on everything that looks as if it could be "Greek" and ensuring that the data held is as correct and consistent as possible. That way when the apparently inevitable happens one can at least have a reasonable set of data to work with.

Now I am not economist and I have to admit that I have not worked this all out and have no idea of the scale, but instinctively I believe I am on the right track.

This could be a very interesting and difficult ride. I just hope bigger brains that mine are thinking about this and can come up with a better solution.