«Ms» mce-anchor«Wilkinson» : We obviously had a look at the Rice Warner work that came out last week. We haven't had an opportunity to talk through the details of that analysis, but we certainly are intending to reach out to Rice Warner or the SMSF Association to have that discussion. It looked on the face of it like they had assumed that we hadn't taken into account any behavioural changes in our costing—which, as I've explained, is not correct at all; we absolutely have. So we were a bit puzzled by that assumption, because we've always been clear in all of our documentation that we take into account behavioural responses. The types of behavioural responses that they talked about in that publication were very similar to the types of behavioural assumptions that we've indeed taken into account. As I said, I'd prefer to take specific quantum questions on notice, but I think I'm comfortable in saying that, in terms of broad orders of magnitude, the sorts of behavioural responses that they have concluded are not that different to the sorts of behavioural responses that we think are reasonably likely. So we took some comfort in the fact that there was a completely independent estimate of what are the likely behavioural responses, and we hadn't had any involvement with that work that was undertaken. Certainly on that part of their analysis, on the ways in which people are likely to respond and what proportion of people are likely to respond, we felt reasonably comfortable.
So, to the extent that there is a difference between where Mr Rice lands and where the PBO lands, it doesn't lie in assumptions around behavioural responses; it lies somewhere else?
mce-anchor«Ms» mce-anchor«Wilkinson» : I guess the thing that I would like to explore with that analysis is a better understanding about, first of all, why they presumed that we hadn't taken into account any behavioural responses and, therefore, what's the baseline and the starting point for that analysis, because, as we've discussed, you've got to start with your baseline as to what's the likely outcome without this policy, and then you make an assumption about the direct impact of the policy and the behavioural responses, and that's how you get what your policy costing is.
