CHAIR —This is not the Oxford debating society or anything of that nature, but I think the committee agreed that since you, Mr Smith, made some remarks about the study that we had presented to us this morning it would be appropriate that we ask the authors of the study to comment on your remarks and then we might all have some questions for you. I point out that we will need to be out of here—at least I will—by 4.30 p.m. If you have any remarks you wish to make, Professor Harding or Professor Warren, would you please take the call.

Prof. Warren —Ann and I very briefly had a chance to talk about the opening comments that Greg made, and we have a small problem with the inference that we misunderstood the government's package. The way we structured the option 1, 2, 3, 4, 5, 6 packages, if you like, was by applying a blanket rule of 3.4 per cent. We knew exactly what we were doing; we did not misunderstand the rule. The point was, if the government did apply this formula and there were variations within the groups of the price indexes, how would that impact across different groups because you have applied a flat rate to all those groups.

So we were quite aware when we did option 3 that we were adding, effectively, with the small difference in terms of the price effect—we had two: we added 1[half ] per cent; the government had 3.4 per cent and we had 3[half ] per cent—and just ran with the 3.4 per cent. The issue then was that, if four turned out to be the case, the approach we took was wrong in terms of the compensation. We acknowledged that in the report and indicated that the approach we took by applying 1.4 per cent in the case of options 4, 5 and so on was not very intelligent because the CPI effect obviously, if you took food out and so on, would be different.

That is why, in a sense, we then dropped to the Bs. The Bs are applying the rule of the CPI, taking out tobacco and housing—and I was slightly confused at one stage whether the inference was that tobacco and housing were in. My understanding of the ANTS document is that they are out of the price index used. Page 16 of the ANTS document indicates that is the approach. As we ran through those B cases, we were essentially modelling the rule that the government implemented, which is the CPI, excluding tobacco and new homes, plus the 1.5 per cent. That is what we did in options 3B, 4B, 5B and in 7. So in a sense we have acknowledged that point, and that is what we have done in modelling through that.

I have a second point, very briefly, in relation to HES and its credibility. We have gone to considerable length to try to distance ourselves from that controversy. That is why we have hypothetical savings ratios and so on. This morning there were some concerns expressed about how you get those hypothetical assumptions, but in a sense we were trying to distance ourselves from a debate that we felt would be fruitless to the point that was being brought home. So that is my response to the two points. Ann has an additional response.

Prof. Harding —I guess I would like to make two points. As I understand it, irrespective of whether one puts tobacco and housing in there or not, the Treasury are saying that the package is fair enough if the compensation is 1.5 per cent above whatever the measured CPI effect is. I think that our position fairly clearly this morning was that, for some groups, we expected the price effect to be one per cent higher than the average CPI effect, so that in effect that 1[half ] per cent buffer would be reduced to half a per cent, and we felt that was getting a bit too close for comfort—that because this was an average effect, once you got to only a half per cent margin, you might expect that some people would be losers whereas other people would be winners to the tune of more than the half a per cent. If they are indicating that the compensation package is not to become more generous, I guess, in the way that we showed you under option 7, for me that raises serious questions about the equity implications of the package.

The second thing is that it sounds to me as though Treasury have just said that the compensation package will be eroded over time. It sounds as though this will happen if average weekly earnings, for example, increase by 10 per cent over the next five years and the CPI increases by, say, seven per cent and then you add the 1[half ] per cent buffer, which would be 8.5 per cent, that would be lower than the AWE increase you would be getting for pensions and, therefore, you would forget the CPI buffer and you would still be indexing to 25 per cent of AWE. The fact that pensioners were now bearing these higher indirect tax burdens would be forgotten and no longer taken into account.

If this is a correct interpretation of what they have just told us is the position of the government, I guess that does call into question the whole food in food out issue, because option 7 was predicated on the assumption that you would try to buttress the compensation package at the bottom end with more generous compensation and that that compensation would be maintained through time. It looks to me now as though Treasury are questioning the validity of both of those assumptions.

Prof. Warren —If I can add to that, the critical observation that comes out of that is the imperative there to formalise the nature of that compensation. Whether it be related to average weekly earnings or a targeted GST credit, there is clearly a need to lock in that compensation in a firm and definite way.

Senator MURRAY —And at a substantial enough level to cater for those who are lowest.

CHAIR —Do you have any quick comment, Mr Smith, or shall I ask the committee to start questions?

Mr Smith —I gather from what has been said that NATSEM confirm that they thought the benefits under options 1 to 6 would be 3.4 per cent and have confirmed that they will not be 3.4 per cent under the price assumptions set down in options 1 to 6. That was my point at the opening and, therefore, the tables are all wrong to the extent that they do not have a matching increase in cost of living and social security benefits. The matching that is provided under the government's policy is not provided in those options. That is my only point.

The other points that were made then by the NATSEM officers were matters of opinion about compensation about which I would not make any remark. It is not a Treasury matter.

CHAIR —The only way we can do that is for me as chair to offer one question each to the committee.

Senator CONROY —I wanted to confirm what you said earlier that pensions will be the same with a GST as they would have been without a GST. I described a banishing and you described a coming together.

Mr Smith —I do not know what your question is. I frankly can only repeat what I have said. It is on the record, that what the government has guaranteed is that at least 1[half ] per cent plus the CPI, inclusive of tobacco and housing effects without any discounting, will be provided as a result of the tax package.

Senator CONROY —But at some point in the future that increase may converge with MTAWE, the 25 per cent level?

Mr Smith —It is possible. I do not know whether it will or it won't. But it is possible that, if in the future in addition another guarantee comes into play, that in addition to having 1.5 per cent above CPI increases in the pension, there may be further increases based on one-quarter of MTAWE.

Senator CONROY —You are saying that at the moment the only commitment is the 25 per cent of MTAWE irrespective of the GST?

Mr Smith —The commitment that I am referring that arises from the GST package is included in the legislation introduced on 2 December last year and is now before the Senate, and it is to increase pensions by 1.5 per cent more than the CPI.

Senator CONROY —But they are not additive. I think those were your words.

Mr Smith —That is to say, the government has not made a commitment that also seeks to add something to the MTAWE guarantee.

Senator FERGUSON —I have more of a comment than a question. If we could have put some pads on Mr Smith and sent him to the West Indies, we would not have lost any tests, I do not think.

CHAIR —That is a comment on your tax ability, rather than your cricketing ability, Mr Smith.

Senator MURRAY —Senator Ferguson will start calling you Bill Laurie. He was notorious for protecting his wicket. Professor Harding, I will turn to your page 11 where you have the estimated number of households within each cameo type. The largest numbers in HES samples—and you probably do not need to look at it—make it possible to disaggregate to establish how different households within those cameos might be affected, depending on what the expenditure patterns are. Therefore, it would be possible—building on your earlier

remarks—for some of those larger sample groups to disaggregate it down to establish those people who are either worse off or who are at the margin and would therefore need greater compensation.

What many people do not understand is that those of us concerned with this issue are not concerned with how well the best off do, but to make sure that the worse off, in fact, get a decent hit. If within a large HES sample somebody for instance got a net benefit of eight per cent, that is not of concern to us, providing the bottom person got 3.5 or four per cent. Do you follow me?

It is possible, is it not, within the modelling approach within certain of these HES sample cameos to at least arrive at a situation where you could identify how many people are either worse off or are at the margin in terms of compensation?

Prof. Harding —I can imagine, for example, for the single persons you might split that into quartiles of single people in the HES or quintiles of single people. I imagine Neil could calculate it and see whether there was a differential price effect for those quintiles. Then you could apply that price effect to the relevant income levels in the cameos. That would be perfectly technically feasible.

Senator MURRAY —It would theoretically be possible—although we have not tested it—for one quintile there to actually have a minus 1.5 consequence of the GST package and another one within the same sample to be a plus 10, depending on their expenditure and income patterns.

Prof. Warren —What you are saying is that the overall price index for the group is low.

Senator MURRAY —That is right; the range.

Prof. Warren —That is correct.

Senator MURRAY —And the difficulty we are experiencing with the Treasury's and the government's approach is that this averaging does not take account of those people who are potentially worse off, the people whom the charities and churches are focused on. That is right, is it not, from that kind of analysis?

Prof. Warren —Yes. If it is a sample size issue, then in the larger sample size issue, obviously, you address your sample size problem and therefore you can investigate that particular issue.

Senator MURRAY —The answer that everyone gets 1[half ] per cent benefit automatically is wrong.

Prof. Warren —In the sense that it is an average.

Senator MURRAY —That is right.

Prof. Warren —As we discussed earlier this morning, an average goes both ways. There are people above it and people below it. In a sense, what you are looking for is how important is that variation within those groups for those people, where that group on average is sailing close to the wind, and that type of issue. If they are quite close to the 3.4, even though we indicate on an average basis that they are still the beneficiary, on average there could be some groups within that who are the losers.

Senator GIBSON —I have no questions.

Senator HARRADINE —I have no questions.

CHAIR —There are no more questions. Do you have any final remarks, Mr Smith?

Mr Smith —I think I have said enough, thank you.

CHAIR —Thank you all for your assistance to the inquiry. It has been appreciated. This hearing now concludes and we will adjourn into private session to prepare our report.

Committee adjourned at 4.28 p.m.