The big question is: do we need a deposit insurance levy at all? There are costs involved in resolving any difficulties with the systemically important banks. Which means the next question is: who should bear this cost? The answer, in New Zealand, is the shareholders first of all, followed by the creditors, in order of insolvency priority. Since there is no deposit insurance in New Zealand, the latter includes depositors, and quite early on in the piece too as they are junior unsecured creditors.Most other countries have come to the conclusion, like Australia, that there needs to be a resolution fund, and that this needs to be paid for up-front, as there will be no easy means of finding such funds when they are needed except through the tax-payer.All this implies that a levy on the banks themselves is necessary. This means the levy will, in practice, be borne by some combination of depositors, borrowers and shareholders. Some countries plan for this levy to cover all depositors, others only the depositors in the banks that could benefit, namely, the systemically important ones.To me this second approach seems fair. The large banks already get an implicit funding subsidy from their too-big-to-fail status compared with the small banks; this will merely make the outcome a little more balanced. (Resolution costs for small banks will be borne by shareholders and then creditors, in order of priority, with the deposit insurer standing in for insured deposits - whose funds would be raised from all the banks.)There are questions about how the money should be held, however. Should it be held by a state institution? Should there be a clear "insurer"? For example, one might argue that the deposit insurer should also manage the funds as it would be  intimately be involved in any resolution as the largest likely creditor. If it were not for inter-agency rivalry in the United States, I imagine that is how it would have gone there. The difficulty about such being an industry body is that it will be these authorities who decide when and how the funds would be used.The key problem for most countries is that they are trying to build up these funds at exactly the same time as they are trying to re-ignite growth in their economies and emerge from the global financial crisis. My view is that we should take this all very slowly. Recapitalisation under Basel III will take until 2019. I would spread out raising the extra resolution funds for the large banks until 2019 to 2022, or possibly to 2025, at 0.05 per cent of deposits a year. We are all presently paranoid about a further crisis. But, really, we are talking about preparation for a very rare event - the failure of a major bank in another crisis, not in the remainder of the present one. And we still have tax-payer dollars available in the interim should there be such a disaster.David Mayes is a member of the advisory panel of