Transcript of Question & Answer Session Panel participation at the Centre for International Finance and Regulation (CIFR) Market and Regulatory Performance Stream Symposium

Okay well thank you very much for the feedback, that's been really good. I guess a key part was as you pointed out that - first Malcolm you pointed out, money isn't free. You pay for a lender of last resort provision and similarly in the actual 2008 experience in Australia the Government actually didn't need to pull out the cheque book and say ‘dear Bank X here's the money that you need’. And thereby taking into account I think with regards to the actual payment of a debt guarantee. So there's two parts to that, one is you can have policies in place that aren't actually needed, and we got over what we would call a crisis compared to some other countries where those policies were needed. And so I guess what we're doing in our paper is we're saying we're setting up a structure where in certain circumstances those policies will be needed. So I guess you could say well look in some ways the actual crisis situation better models the US or Ireland or Iceland, rather than modelling what actually happened here in Australia. But what we're doing is we're taking the step back and we're saying okay well assuming that there is a possibility in Australia that those funds will be needed at some time – we didn't need them in 2008, thank goodness – but they may be needed in the future. What are the implications of that for banking structure and for interest rates today? So I guess that's what we're looking at there.

On the payment for the guarantee, yeah I was actually thinking about that when I wasn't paying attention to – I apologise to one of the people, one of my fellow presenters, because that was raised during the question time and I thought, yeah, got to be able to work that out and that is an interesting question. I think it's actually fairly easy to put into our model, of having you know in a sense an idea of the right payment and working out what that will do. It will still have implications on the banking structure because it will reduce the number of banks, in equilibrium it will have a competitive effect. Now again we may say that is the cost that we have to have in place to enable the banking system to cover a crisis but we need to understand what those costs are.

Just another comment on that. The reason I raised the concept of self-fulfilling runs is because I think it helps to work against your fundamental conclusion. Your point is if you provide any sort of government support or prospect of support or guarantee – lender of last resort, I think falls into that category – you're protecting banks against risk and therefore encouraging them to take more risk and therefore potentially making a crisis more likely. But my point is that the existence of those support mechanisms also has the beneficial effect of reducing the likelihood of a self-fulfilling run on a bank.

Again not saying these don't have benefits and obviously you can have a line of credit for the banks which stops a run on the banks, that's great. Now if it was a zero probability of that line of credit ever being drawn on, then there would be no implications for the banking system. We've got a positive probability that that line of credit will actually be required and that's where it feeds back in and yes, no troubles bringing in what if there was a limit on sovereign support? What if you went down Iceland's approach where you said well we'll protect the domestic depositors, tough luck if you're a foreigner? All of that can be built into the structure.