Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Economics Legislation Committee
Treasury Laws Amendment (Personal Income Tax Plan) Bill 2018

EWING, Mr Robert Joseph, Principal Adviser, Tax Analysis Division, The Treasury

MRAKOVCIC, Ms Maryanne, Deputy Secretary, Revenue Group, The Treasury

PURVIS-SMITH, Ms Marisa, Division Head, Individuals and Indirect Tax Division, The Treasury

CHAIR: Welcome. I remind officials that the Senate has resolved that an officer of a department of the Commonwealth or of a state or territory shall not be asked to give opinions on matters of policy and shall be given reasonable opportunity to refer questions asked of the officer to superior officers or to a minister. This resolution prohibits only questions asking for opinions on matters of policy and does not preclude questions asking for explanations of policies or factual questions about when and how policies were adopted. Thank you all for appearing before the committee today. I invite you to make a brief opening statement, should you wish to do so.

Ms Mrakovcic : We have made a submission to the committee and some supplementary material was provided at last week's hearings. We have nothing further to add.

CHAIR: Can Treasury please outline for the committee who currently pays personal income tax, what proportion of tax is paid by those on the top marginal tax rate, how much that is in dollar terms, what proportion is paid by the top one per cent of taxpayers and those on lower marginal tax rates, what percentage of taxpayers they make up and how much tax they pay? Perhaps give us a broader picture of the demographic breakdown of personal income taxpayers now.

Ms Mrakovcic : We can outline for you that, according to the 2015-16 tax stats, which I believe are the latest year available that we're quoting, the top one per cent of taxpayers paid around 16.9 per cent of all tax paid, the top five per cent paid around 33 per cent, the top 10 per cent paid around 44.9 per cent, the top 25 per cent paid around 67 per cent and the top 50 per cent paid around 87.5 per cent.

CHAIR: Do you have that in dollar terms?

Ms Purvis-Smith : We don't have the percentiles that we just went through in dollar terms, unfortunately, but we can take that on notice for you if you like.

CHAIR: If you could, I think that would be very handy.

Senator KETTER: Do you have a figure as to what wage or income level the top 10 per cent, say, represents?

Mr Ewing : No, we would have to take that on notice.

CHAIR: Can you take me through data on the number of people who pay no net tax after transfer payments?

Mr Ewing : Our estimate is that in 2015-16 about 46 per cent of Australian adults did not pay any personal tax. I think the question you're asking then goes to the transfer system and the net point of that, on which I don't think I have an up-to-date figure for you, so I'm afraid I'll have to take that on notice as well.

CHAIR: They paid no tax at all, or no net tax?

Mr Ewing : Around 46 per cent of Australian adults did not pay any personal income tax in 2015-16. That's based on ATO tax statistics and ABS demographic statistics.

CHAIR: So we don't have the data for after transfer payments?

Mr Ewing : No, it's a slightly more complicated calculation.

Ms Purvis-Smith : Something in one of the budget documents might help you. While it's not a time series, in 2015-16 around 40 per cent of Australian households, rather than individuals, received more in government payments than they paid in income taxes. It's a slightly different concept than what Mr Ewing was referring to.

CHAIR: How does Australia's top marginal tax rate compare to those of other advanced economies like Canada, New Zealand or the UK? Where do their thresholds cut in, more importantly, on multiples of earnings?

Ms Mrakovcic : I make the general observation that among OECD countries when you look at Australia's top marginal tax rate we are relatively high in the rankings. We have a relatively high top tax rate and it kicks in at a relatively low multiple of incomes. On both of those statistics in the ranking of OECD countries we would be relatively high, around that 10 to 15 mark. Ms Purvis-Smith, do you have anything to add?

Ms Purvis-Smith : At the moment our top marginal tax rate cuts in at 2.2 times full-time average weekly ordinary time earnings, compared with around four times in Canada and the UK and eight times in the US.

CHAIR: The $180,000 threshold has been around for some time. Do you have what multiple of average earnings it was when originally introduced?

Ms Purvis-Smith : Bear with me. I should have that for you.

Mr Ewing : I believe I have won the race to find it in our folders! The $180,000 threshold was introduced in the 2008-09 income year. At that time, AWOTE was $61,158, so it was a little bit under three times AWOTE in that year.

Ms Mrakovcic : We're going to, in turn, beat you and say that, according to what we've got in front of us, it was 2.9 in 2008-09.

CHAIR: And now it's 2.2?

Ms Mrakovcic : As at 2016-17, it's 2.2.

CHAIR: When the highest marginal tax rate threshold increases to $200,000 under the government's proposed Personal Income Tax Plan, do we have an estimate of what the multiple of average weekly earnings would be at that stage?

Ms Mrakovcic : Do you mean with no change in the system or under the personal tax plan?

CHAIR: Under the personal tax plan, the highest marginal tax rate threshold will go to $200,000 in 2023-24, am I right?

Ms Purvis-Smith : In 2024-25.

CHAIR: Thank you. Do you have an estimate of what average weekly earnings will be in that year so that we can work out what the multiple will be, and how will it compare?

Ms Purvis-Smith : I do have an estimate of the multiples.

CHAIR: Yes, you can only estimate it.

Ms Purvis-Smith : Under the government's plan, the ratio of full-time average weekly ordinary time earnings, as a multiple of the top threshold, will be around 1.9 times.

CHAIR: That makes it sound like we're becoming less progressive—a flatter system, certainly. I should probably qualify that. I would have thought that we would try to maintain somewhere between 2.9, which is where the $180,000 kicked in 2008-09, or 2.2—now it's actually coming down. The highest marginal tax rate kicks in at a lower level compared to average weekly earnings.

Ms Mrakovcic : That's correct. In a sense, it is expected to go from 2.2 to 1.9; in the counterfactual, it would go to 1.7.

CHAIR: That's if there is no change?

Ms Purvis-Smith : That's correct, Senator.

CHAIR: Some witnesses today suggested that the personal income tax package would make the system more progressive or less progressive or would have no change whatsoever on the progressivity of the personal income tax system. What is the Treasury's perspective on that?

Ms Mrakovcic : I make the observation that progressivity is a complicated concept and there is no single indicator that will somehow or other give you a magical answer to the overall progressivity of the tax system. If you look at some of the statistics around what happens to AWOTE, if you look at what happens to the proportion of tax paid by people in that top tax bracket and if you compare that to where things are at the moment, in overall terms we anticipate that there'd probably be very little change in overall progressivity of the tax system.

CHAIR: So all of that arguing we did today was for nought, if that's the case. All right. Can I ask about bracket creep? Perhaps the Treasury can explain to the committee exactly why bracket creep is such a problem.

Ms Mrakovcic : Bracket creep refers to the fact that taxpayers' face higher average and marginal tax rates over time, even if their income is only increasing by inflation. Essentially, because your income is rising, there are two scenarios. One is where even if you do not actually have a higher marginal tax rate, just by virtue of the fact that more of your income is at that higher marginal rate you're currently on, your average rate will go up. Of course, that would be magnified if, in fact, you do cross a threshold, when you will also face a higher marginal tax rate.

CHAIR: So bracket creep impacts the progressivity of the system?

Ms Mrakovcic : Certainly, you would anticipate that bracket creep would impact on the progressivity of the tax system. Those impacts will, of course, vary year-to-year and then over the very, very long term. You could imagine a scenario where, in fact, if you don't have any change, more and more of your population is actually going to come into those higher tax brackets.

CHAIR: I think it's important, for the record, that we ask Treasury just to give a very brief overview of the government's personal income tax plan—the first step, the second step and the third step—and then I will ask you a couple of questions about those individual steps.

Ms Mrakovcic : Certainly. I might ask Ms Purvis-Smith to do that.

Ms Purvis-Smith : Step 1 introduces a new tax offset called the low- and middle-income tax offset. I won't go into the minutia of the detail, but it tapers in up to a maximum of $530. Individual taxpayers receive the maximum amount of the offset at $530 between the taxable incomes of $48,000 and $90,000 and then it tapers out. It phases down and by the time you get to just over $125,000, no tax offset is received.

It is a non-refundable tax offset, which means that it reduces the tax liability that you have. Another way of putting that is that you must pay tax in order to receive the offset. The offset is also received at the end of the tax year—the financial year. It's received as a lump sum on assessment once an individual puts in their tax return to the ATO—when they lodge their tax return.

That starts from the 2018-19 income year; it starts on 1 July 2018 but it's not received until—

CHAIR: The first tax offset should, potentially, be received on 1 July 2019?

Ms Purvis-Smith : That's right—when you put your tax return in for that income year—

CHAIR: But in the next financial year?

Ms Purvis-Smith : That's right. On 1 July 2018 also—this is part of step 2—there is a threshold change. The threshold of $87,000 is increasing to $90,000. Then from 1 July 2022, the low- and middle-income tax offset applies for four income years. So from 1 July 2022, the low- and middle-income tax offset is removed and the benefit being provided is provided through a different mechanism. That mechanism is twofold. It is moving the $37,000 threshold to $41,000 and also increasing the low-income tax offset by $200 to $645. That provides a maximum amount of $540 for individuals.

In addition, from 1 July 2022, and also as part of step 2, the top threshold of the 32½ per cent tax bracket will be increased from $90,000 to $120,000. Step 3 is the removal of the 37 per cent tax bracket. That means that the $120,000 threshold will increase to $200,000, thereby removing the 37 per cent tax bracket completely.

CHAIR: Right. Your submission explained the difference between marginal tax rates and average tax rates, and I think that's come up a few times today. So is it true that under the government's plan, by the end of step 3—that third step in 2024-25—a person earning $200,000 would face a much higher average tax rate than an individual earning $41,000? That came up with one of the groups today—I can't remember which one it was. The ACTU, I think it was.

Ms Purvis-Smith : That is correct. Being on the same marginal rate, the way the tax system is structured means that even though you might have two people within that marginal tax rate you can be paying, and you will be paying, quite different average tax rates within that range. Someone on $42,000, for example, will be paying a much lower average tax rate than someone on $180,000, for example.

Senator KETTER: That's just because of the impact of the tax free threshold and the amount is lower—

Ms Purvis-Smith : That is part of the way we have a progressive tax system. But the impact of the tax free threshold is that, as your income increases, it is a much lower proportion of your income, so your average tax rate increases over that income range.

CHAIR: One of the things the Treasurer was saying on budget night was that for the vast majority of people—I can't remember the exact percentage he used—over their working life their tax rates would not necessarily change, if the 37 per cent marginal tax rate were abolished and the 32.5 per cent marginal tax rate applied between $40,000 and $200,000. Is that correct? Is that Treasury's understanding, as well?

Ms Mrakovcic : Yes.

CHAIR: Good, I'm glad the Treasurer was not mistaken when he said that! What would be—and you might have to get your calculators out—the cumulative percentage reduction in tax paid under the government's plan, by 2024-25, for an income earner who earns, say, $50,000 or $90,000 or $200,000? Is that a bit too hard? You can perhaps take it on notice.

Ms Mrakovcic : We'll take that one on notice.

CHAIR: I thought I might be asking a little bit too much for you to calculate on to the spot. I will turn the questions over to the deputy chair, but I might come back to you with a couple more.

Senator KETTER: Have you had a chance to review the PBO response, and do you have any introductory comments you'd like to make about that?

Ms Mrakovcic : We have had an opportunity to have a quick look at it.

Senator KETTER: Would you like to comment?

Ms Mrakovcic : I don't think we have any comments.

Senator KETTER: This analysis confirms that it is possible to do year-on-year costings. I want to double-check that Treasury has done year-on-year costings in relation to the projections?

Ms Mrakovcic : I think we traversed this ground last week—

Senator KETTER: A simple yes or no is fine?

Ms Mrakovcic : I think we've got nothing to add to the comments we made on this matter last week.

Senator KETTER: I think you indicated those costings had been done. Is it the Treasurer's office that's denying the Senate the opportunity to scrutinise these year-on-year costings?

Ms Mrakovcic : We made the observation last week on this matter—you'll recall I indicated that for us the issue is that the further out you go, the more there are inherent uncertainties in those estimates. I also indicated that you have to balance these things. We're very conscious of the fact that the cost over the forward estimates is $13.4 billion, and it's set out in the budget measures. But, obviously, there are steps that occur outside the forward estimates. We thought that it was important to be able to give a sense of the medium- to longer-term cost of it, or 10-year cost, as a total to give a sense of the ballpark figures that might be talked about, in terms of all aspects of the package. But, by the same token, because there is that inherent uncertainty, we have concerns around the fact that you can actually end up having what we think is an undue significance attached to those kind of annual estimates that are produced. We have concerns around those annual estimates being used with pinpoint accuracy. I think that there was quite a bit of information that was provided last week, certainly in terms of stepping out the costs of each of those components. I think that there was material provided to give a sense of what's happening in the overall package but, in terms of those annual estimates, as we said last week, we are far more comfortable talking in aggregate terms about the cost of the package and those component parts than we are for the year-to-year figures to be out there.

Senator KETTER: What do you say about the costing the PBO has done for the total to 2028-29 of $143.95 billion?

Ms Mrakovcic : We have no comments to make on the PBO costings. They've obviously produced their numbers and we've produced our numbers.

Mr Ewing : The methodology that I have in front of me says that the PBO have calibrated their costing to the total figure that the Treasurer provided in parliament on 23 May, which I believe was $143.95 billion.

Senator KETTER: So we've got the total package cost. You've given us the first four years. You're not prepared to give us figures for the intervening years, even if you put caveats on them and tell us what the assumptions are and your degree of confidence. Surely this is where we need to have information in the public domain to consider the package and its impact? Particularly over the medium term if we have the potential for downturn in the economy and what impact that might have, these are reasonable public questions that should be answered.

Ms Mrakovcic : The secretary addressed this issue last week in estimates, I think I made some observations last week on this issue and I've also addressed the issue just now. I really don't have anything to add to those comments.

Senator KETTER: Also, I just want to confirm, again, that there was no gender analysis done in relation to the impact of the particular measure.

Ms Mrakovcic : That's correct.

Senator KETTER: Did the government instruct Treasury not to carry out a gender analysis?

Ms Mrakovcic : I think it's fair to say that we didn't do gender analysis because, as the secretary again pointed out last week, we actually see the tax system as gender-neutral. For us, a man earning the same amount as a woman pays the same amount of tax as a woman. They have the same individual tax outcomes. That's not to say that, depending on the distribution of income in totality among women against males, those won't lead to different aggregate outcomes for men and women, but that's going to reflect the distribution of income in society more generally. As we all know, those issues go to a much broader set of factors than the tax system. The tax system is not, in itself, influencing that.

As far as the government's personal tax plan goes, it was designed, in the government's words, to deliver lower taxes, protect against bracket creep and lead to a simpler system. It did not target gender issues. Having listened to some of the people appearing in front of the committee earlier today, the only more general point I'd also make is that people make the observation that the tax system is complicated and I think it's important not to impose far too many objectives on our tax system.

Senator KETTER: You don't think that we should be looking at the impact of budgetary measures on things like the gender pay gap and income levels?

Ms Mrakovcic : As I said, issues around the distribution of income for women and for men and the implications—what that says about us as a society and whether those things should be an important focus of policy—are for government to consider. I'm simply making the statement that the outcomes that you get in the tax system if you are a male or a female earning the same amount of money are identical, and in that sense we didn't undertake gender analysis as part of the personal tax plan.

Senator KETTER: I'm going to just step through the components of the package. Can you confirm what the cost is over the medium term of the measures that are commencing on 1 July 2018?

Ms Mrakovcic : Based on the material that we provided last week, I see in table 2 that the medium-term cost of step 1, the low- and middle-income tax offset, is $16 billion. If you also want to include the first component of step 2 commencing 1 July 2018, the increase in the top threshold of the 32.5 per cent tax bracket from $87,000 to $90,000 is minus $6 billion, or a $6 billion cost.

Senator KETTER: The split between men and women is $12.5 billion and $9.82 billion.

Mr Ewing : Those are the numbers on the PBO's piece of paper here, but we can't comment on how they've arrived at those and whether they're accurate calculations.

Senator KETTER: And the measures commencing on 1 July 2022?

Ms Mrakovcic : Again, referencing table 2, it's the last three components in step 2. The first of which is increasing the low-income tax offset, which is costed at minus $3 billion. Increasing the top threshold of the 19 per cent tax bracket from $37,000 to $41,000 is minus $41 billion. Commencing 1 July 2022, increasing the top threshold of the 32.5 per cent tax bracket from $90,000 to $120,000 is minus $36 billion.

Senator KETTER: Again there is a gender impact, but again you're not going to comment—

Ms Mrakovcic : Senator, we're not going to comment on any of those gender impacts. As I said, we didn't do the gender analysis and we can't comment on another body's analysis.

Senator KETTER: For the measures commencing 1 July 2024?

Ms Mrakovcic : That would be the $33 billion and the $9 billion.

Senator KETTER: Sorry?

Ms Mrakovcic : It is $33 billion for the increase in the 32.5 per cent tax bracket from $120,000 to $200,000, and increasing the threshold for the 45 per cent tax bracket from $180,000 to $200,000 costs $9 billion.

Senator KETTER: And this is where the PBO says that the gender breakup is $30.35 billion for men and $11.25 billion for women, which you're not prepared to comment on. Can you confirm that the cost of the stage 3 tax cuts grows at about 12 per cent per year?

Ms Mrakovcic : We can't comment on that.

Senator KETTER: Can you at least comment that this is the fastest-growing stage of the entire package?

Ms Mrakovcic : We can't comment on that.

Senator KETTER: So you're not prepared to say anything about the growth rate of stage 3?

Ms Mrakovcic : We have nothing to add to the information that was released last week.

Senator KETTER: Do you agree that year-on-year estimates outside the forward estimates are rubbery?

Ms Mrakovcic : I don't know that I would ever refer to anything as 'rubbery'. I would simply make the point that I have already made: obviously, when you're undertaking projections, the further out you go the greater the uncertainty that attaches to those. We wax lyrical on this subject in statement 8, where we devote quite a bit of time to talking about the sensitivity of our estimates to changes in parameters. From our point of view, we always do the best forecasts and projections that we can, but, because we're preparing them, we are very conscious of the fact that those uncertainties get magnified and that you need to keep that in mind even as you are producing results. Are they the best numbers that we are capable of pulling together? Yes. Is there uncertainty attached to them? Yes. Do those uncertainties manifest themselves more greatly the further out you go? By definition, but we deal with those. As I said, we came to the judgement that we felt comfortable talking about the medium-term totality of the impact, and we're comfortable with that.

Senator KETTER: I still grapple with that concept, that we can be confident about the $143.95 billion figure and yet be so uncertain about putting out the year-on-year costs beyond the forward estimates?

Ms Mrakovcic : I think these are all relative terms—

Senator KETTER: But surely there's a spreadsheet that you've got that provides for line items for each of those years leading up to the end of the period? The final figure of $143.95 billion is a summation of those annual figures. So you're confident about the final figure, but you won't tell us the intervening figures?

Ms Mrakovcic : I think that what we have said is that the more aggregated the level at which you are operating, you can have perhaps a bit more confidence about those numbers. Last week I was here when Mr Ray, I think, took you through the forecasting process and the preparation of the projections. You'll recall that there was a conversation around how there's a certain methodology that is undertaken through the forecast period, and then there is a consideration of a period of time over which the output gap is closing and then movement after that output gap has closed for further methodological change. The point that was being raised was that there are assumptions made around how quickly that output gap closes. So your output gap may close a little bit faster, it may close a little bit slower.

Senator KETTER: Of course, and if you're transparent—

Ms Mrakovcic : Those changes basically may mean that the year-to-year figures may move around a bit, but does that change your sense of what you may be looking at in the overall medium term? I think I'd be less worried in that sense.

Senator KETTER: But isn't it better to be transparent and put on the table what the projections or the assumptions are in respect of that particular assumption you're referring to?

Ms Mrakovcic : You've asked us this any number of times, and I can only go back to the observations that I made last week and I'm repeating this week: we are comfortable releasing the figures that we have, but we did not feel comfortable releasing the annual figures. We've got nothing to add—I really can't add anything further to that conversation. I've tried to explain and I can't do more.

Senator KETTER: I understand what you're saying, but we're being asked to vote on a set of tax cuts that don't come into existence until beyond the forward estimates, and there is a question of having confidence about how those figures have been put together. You're prepared to give us a total package but not what might happen year-on-year beyond the budget, and I think that's a very great concern. This doesn't happen too often, where you've got legislated tax cuts outside of the period of the budget.

Ms Mrakovcic : As I said, I think that there is a fair amount of information that has been provided, including in table 2, around each of those component steps, and we have done our best to provide that ballpark figure, a totality figure. That's it.

Senator KETTER: Let me move back to some of the questions the chair raised. Just in terms of international comparisons of the tax burden or the tax liability—I'm not sure what the correct terminology is—of income taxpayers in Australia compared to other countries, how does one make a useful comparison, given that we have a fairly significant tax-free threshold. How does that impact? Are there any statistics that Treasury would rely on to look at what the level of the tax burden on Australian income taxpayers is?

Ms Mrakovcic : I guess it depends on what you mean by the 'tax burden'.

Senator KETTER: The average tax rate goes some way to picking up some of those issues. If Treasury is looking at this issue, what would be the set of data that you would go to, to look?

Ms Mrakovcic : When we've talked about average tax rates, when we've talked about that top marginal tax rate, when we've talked about the threshold at which it kicks in and when we've talked about what proportion of taxpayers pay different proportions of tax, I would say it's the totality of all of those things that essentially try to paint a picture of your overall tax system. It's not that I would necessarily jump on one summary statistic.

Senator KETTER: There is no OECD dataset that we can go to?

Ms Mrakovcic : Certainly, in terms of some of the OECD analysis that I've seen, there tends to be a focus on what is that top marginal tax rate—

Senator KETTER: But it doesn't tell us the overall picture, does it? It's only one part of the picture.

Ms Mrakovcic : I was going to say 'and at what level of income, and at a relatively low level of income it steps in'. That would probably be an important aspect. I guess you could look at average tax rates as well.

Senator KETTER: There is a dataset out there? The OECD puts out those sorts of figures?

Ms Mrakovcic : I don't know precisely what the OECD puts out. We certainly look at those things and have an idea of what average tax rates are.

Senator KETTER: I'm interested in what is a valid exercise of looking at the level of average tax or overall tax that is paid. I know it is a highly complex matter and there are transfer payments as well that some might look at.

Ms Mrakovcic : Senator, I would say that you could start at a very high level and say, 'What is your tax-to-GDP ratio?' And then you might say, 'In terms of your overall tax-to-GDP ratio, how important are personal income taxes to that, how important are corporate taxes to that and how important are other tax bases to that?'

Senator KETTER: I'm trying to get as much of a sense of the individual impact as I can.

Ms Mrakovcic : And then, essentially, as I said, you would look at what kind of span your marginal tax rates are and you would look at the thresholds that they would come in at. And you would obviously have an eye to what was happening to your average and marginal tax rates. I don't know whether Ms Purvis-Smith or Mr Ewing—

Senator KETTER: If we looked at, for example, average weekly ordinary time earnings across different jurisdictions and identify what the amount of tax paid is. I'll leave that with you. If you could come back to me on that, I would be interested in it.

Ms Purvis-Smith : We can take it on notice, Senator.

Senator KETTER: If I could just move to some of the assumptions. We've heard some criticism today about some of the budget assumptions. Firstly, productivity growth is assumed to be at 1.6 per cent. Is that over the forward estimates and the medium term?

Ms Mrakovcic : We would probably rely on our macro-economic group colleagues on those kinds of issues. I don't know if they make assumptions around that productivity growth in the forward estimates. It's probably best if we take that question on notice.

Senator KETTER: You might want to take some of these on notice. I'd be interested in more information on the migration figures. I did talk to ISA about these assumptions. What are the migration figures that are implied in the budget over the forward and medium term compared with the level of migration that's occurred over the last five years?

Mr Ewing : We'd have to take those on notice as well.

Senator KETTER: Okay; and the assumption about how easily those migrants would obtain jobs. The figure for wages growth is something we've talked about a fair bit. Is that an area that you might be more familiar with? What's the historical outcome for wages growth in the past five years?

Ms Mrakovcic : We'd have to take that on notice.

Senator KETTER: What about the terms of trade?

Ms Mrakovcic : In terms of the forecast?

Senator KETTER: Yes.

Ms Mrakovcic : My understanding is that the terms of trade are set out in table 1, which is on page 2-6. It sets out the forecasts for the terms of trade. My understanding is also that there is some commentary in the statement on the economic outlook that indicates that the terms of trade beyond that forecast period over the kind of projection period of the next few years is expected to remain relatively flat. So I think there is quite a bit of information on the terms of trade set out in that chapter.

Senator KETTER: Did you hear the testimony from Dr Anthony?

Ms Mrakovcic : I did. I was struggling to understand it relative to the forecasts that were set out for the terms of trade in that table. You'll see that in 2018-19, they're expected to decline by 5¼ per cent and 2¼ per cent the following year. My recollection is that he seemed to indicate that there was an assumption that the terms of trade were not going to decline very much or going to stay steady. Perhaps you can remind me of what he said.

Senator KETTER: He was saying that the terms of trade are currently about 47 per cent above the long-term postwar average. I don't know that he went to this specifically, but in previous interviews he said that the terms of trade are being driven by Chinese credit creation monetary policy and the Chinese are tightening at the moment. So he was pointing to the downside risk there.

Ms Mrakovcic : During the forecast period, we take a full range of factors into account and we come to our best view. I'll certainly seek more information from our Macroeconomic Group colleagues, but I would imagine that if they were here they would say that the forecast for the terms of trade set out in that table would represent their best judgement as to the expected change in the terms of trade.

Senator KETTER: I'm going to ask you about your views about the economic and geopolitical risks in the international environment going forward. Just before I ask you, during estimates I had the opportunity to ask Mr Fraser about that and I asked him whether there are any more risks now than there were a year or two ago in relation to the economic outlook, and his response was unequivocally 'Yes, with the geopolitical risks,' and he said, 'That's very sad.' We're going into a period of heightened uncertainty. What do you say about that?

Ms Mrakovcic : I have nothing to add to the comments of either the secretary or Mr Ray, who's the head of Macroeconomic Group.

Senator KETTER: I'll now go back to the package itself and try to understand the policy objectives of the package. I think you mentioned that there was a simplified tax system. Could you take us through those again?

Ms Mrakovcic : I think that in broad terms it was lower taxes; to protect against bracket creep; and a simpler system.

Senator KETTER: No sorts of objectives in terms of workforce participation, consumer spending or any of those sorts of things?

Ms Mrakovcic : Well, I think that the government's objectives as per its Personal Income Tax Plan are set out in what the government has said about its package. That's not to say it doesn't have those impacts or it wouldn't impact in those ways, but—

Senator KETTER: It doesn't set out to do those?

Ms Mrakovcic : the government has set out, certainly, the objectives it set itself in terms of the personal tax plan. But I think that those were certainly issues that the government would have been aware of, and they are captured in some of the language around reward for effort and incentives to take on extra hours to work et cetera. So I think that, if you go into the detail around where it explains those issues, it does traverse some of that ground. And certainly the budget documents also make reference to the forecast taking into account the impacts of the personal tax plan on consumption.

Senator KETTER: Does each of the three stages have a different objective? For example, does stage 1 have a particular policy objective?

Ms Mrakovcic : I think that step 1 in terms of the low- and middle-income tax offset was to provide relief to that lower and middle-income range. I don't know that I could really add much more to that.

Senator KETTER: Okay, stage 2—what's the policy objective there?

Ms Mrakovcic : I think that, if you look at some of those elements, it, in a sense, locks in structurally those components of stage 1, and it puts in place some of those mechanisms around starting to protect from bracket creep.

And then, with stage 3, you have that removal of that tax bracket and, if you like, that further protection from bracket creep and also that simplification through the reduction in the number of brackets. Ms Purvis-Smith, do you want to add anything to that?

Ms Purvis-Smith : No, I think Ms Mrakovcic has described that well. The one thing I would say—and I think the government and the Treasurer have made this point—is that they are linked. Each stage is linked. For example, in particular, stage 1, or the LMITO, should be seen in conjunction with stage 2, as the LMITO is not an ongoing offset. It does stop after four years. So there are quite a lot of linkages between each of the stages.

Senator KETTER: Just on the LMITO: what's the policy rationale for making the LMITO available to everyone earning above about $40,000 per year rather than tapering off between $90,000 and $125,000?

Ms Purvis-Smith : Sorry, can I clarify that? Are you asking what's the rationale for—

Senator KETTER: Making the LMITO available to everyone earning above $40,000 rather than a tapering off between $90,000 and $125,000.

Ms Purvis-Smith : The LMITO does taper off. I may be misinterpreting your question, Senator. The LMITO does taper off between $90,000 and approximately $125,000 and phases in as well. The maximum LMITO is provided to taxable incomes between $48,000 and $90,000, and then it phases in as well, so it's not provided to everybody, all individuals, above $40,000. But did I misinterpret your question?

Senator KETTER: So you're abolishing the LMITO but making an equivalent benefit available to everyone above a $40,000 income by bracket and rate changes?

Ms Purvis-Smith : My apologies. You're referencing stage 2 of the plan?

Senator KETTER: Yes, I'm sorry. I was talking about stage 2, yes.

Ms Mrakovcic : That's when the LMITO phases out, and it's replaced with stage 2.

Ms Purvis-Smith : LMITO stops—you're correct—and it's in place for four years. Again, all decisions around what to do and what to do in the thresholds are government decisions. Moving the $37,000 threshold to a $41,000 threshold provides a similar amount of relief to the LMITO, but on a more structural basis and an ongoing basis. The increase in the LITO is also to make sure that the benefits of the tax relief from LMITO continue through in an ongoing way.

Senator KETTER: Stage 2 seems to have different policy measures lumped together. They are very different policy measures—the changes to the thresholds as well as that sort of translation of the benefit of the LMITO into the scales. Can you tell me why they were put together. Who initiated the construction of stage 2?

Ms Purvis-Smith : I think all I can say is that everything that's in the budget is a government decision.

Senator KETTER: Finally, what's the policy objective for stage 3?

Ms Mrakovcic : I think the Treasurer has articulated that. It removes that 37 per cent tax bracket and it basically ensures that you do have 94 per cent of taxpayers essentially having a marginal tax rate no higher than 32.5 per cent.

Senator KETTER: Are there any assumptions as far as outcomes in relation to propensity of people to work additional hours or their level of participation? Are you expecting a particular level of outcomes from stage 3 in terms of the impact on people being more prepared to work additional hours or more people participating in the workforce?

Ms Mrakovcic : Let me separate that into two questions. We haven't modelled, or don't have, a quantitative impact on those issues. But there is a body of evidence, including from the OECD and others, that would indicate that you would anticipate that these kinds of changes would definitely have implications on incentives to work, to save and to invest. Essentially, all the changes in marginal tax rates would be anticipated to have that kind of effect.

Senator KETTER: But all the evidence suggests that people on higher incomes are less likely to change their behaviour in terms of participation as a result of a change to the taxes.

Ms Mrakovcic : There has been a lot of commentary on this issue. Obviously, different commentators have different views on those incentives. I have been listening over the course of today, and I have heard a number of comments around those different elasticities—whether it's at the top end, whether it's at the bottom end and where you want to provide those incentives. I guess I'm just making the general observation that I would anticipate that lower tax rates would actually provide an incentive to work more hours and possibly to participate.

Senator KETTER: Even at those levels of income?

Ms Mrakovcic : The only thing I probably would mention is that I'm certainly aware of some OECD analysis which would seem to suggest that average tax rates perhaps influence participation and lower marginal tax rates probably provide more of an incentive to increase your hours worked. Of course, there is a broader set of issues I should have mentioned around the whole issue of the interaction of the tax and transfer system and effective marginal tax rates. Certainly, when you start looking at participation, I think it's important to have a complete understanding of the way the entire tax and transfer system works.

Senator KETTER: But there has been evidence earlier today about this issue of higher-income individuals not being particularly incentivised by tax cuts to work more or participate.

Ms Mrakovcic : Certainly I heard that. I would simply make the observation that, just generally, I would anticipate that higher marginal tax rates would probably lead one to make decisions around one's incentive to work. Now, whether or not the responsiveness is more or less, as I said, I've seen varying evidence on this issue. But the fact that marginal tax rates incentivise behaviour is clear.

Senator KETTER: And what do you say about the 10,000 millionaires rushing to take advantage of our tax rates? Did you hear that evidence?

CHAIR: That was not at all what the evidence was. You twisted those words, Senator Ketter. I'm very glad my ears pricked up on that one.

Ms Mrakovcic : I think I may have missed that particular piece. I will say I wasn't tuned to it all day.

Senator KETTER: We've heard evidence—I think it was from the Grattan Institute, was it?

CHAIR: No, I think it might have been from The Australia Institute.

Senator KETTER: The figures are that there has been an influx of millionaires into Australia. Are you familiar with that? Ten thousand was the number that was mentioned.

Ms Mrakovcic : No, I'm not aware of that.

CHAIR: Apparently, the Chinese are just thrilled with our tax rates, so they're all coming here in droves! I think it's stretching a bit of a long bow.

Senator KETTER: I think it was in fact the Grattan Institute. Mr Daley made that comment. Could you take that on notice for me?

Ms Mrakovcic : Sure. I just want to make sure I understand the proposition. They're coming for our high marginal tax rates?

CHAIR: Apparently we have had 1,000 new millionaires move to Australia. It has nothing to do with their income or anything like that; it's just millionaires. I think it's an interesting statistic.

Senator KETTER: Was it 1,000 or 10,000?

CHAIR: I can't remember what the number was, but whatever it was—

Senator KETTER: It was interesting.

Ms Mrakovcic : Sorry, Senator, what would you like us to get back to you on?

Senator KETTER: I'd like you to verify that for me. Has that actually happened?

Ms Mrakovcic : That 1,000 millionaires?

Senator KETTER: How about we come back to you with the Hansard.

Ms Mrakovcic : If you could provide us with a very specific question, we'll endeavour to get back to you.

Senator KETTER: Okay. Now, in finishing up, I just want to give you a final opportunity to say anything about the gender distributional impacts of the budget, particularly in relation to the income tax cuts. This is your opportunity to say something. I understand that earlier today the Treasurer said that it was ridiculous to suggest that the stage 3 of the personal tax cuts would benefit men more than women. He said it was ridiculous. There must be some evidence or information underpinning that comment.

Ms Mrakovcic : I have nothing to add to my earlier comments on this.

Senator KETTER: So where would the Treasurer be getting that information from?

Ms Mrakovcic : I don't have anything to add to what I said earlier.

Senator KETTER: Could you take on notice—

Ms Mrakovcic : I'm happy to take that on notice.

Senator KETTER: what evidence the Treasurer is relying on to make that comment?

CHAIR: It sounds like an opinion to me.

Senator KETTER: No, I'm asking for the evidence base of the Treasurer's pronouncement. In terms of geographical distributional impacts, do you have anything to tell us about that?

Ms Mrakovcic : No.

Senator KETTER: So neither gender nor geographical distributional impacts were considered in the consideration of the preparation of this budget measure.

Ms Mrakovcic : That's correct.

Senator KETTER: Okay. Thank you very much.

CHAIR: Senator Storer?

Senator STORER: Would you have any information on the fiscal impact of extending the low and medium income tax offset beyond 2021-22? Would it be a similar level if it were extended to the end of 2028-29 period?

Ms Mrakovcic : We'd have to take that on notice, I believe. Mr Ewing, do you have anything to add?

Mr Ewing : We'd have to take any specifics on notice, but I don't think that there will be anything we can supply. It's, I suppose, not necessarily a straightforward question, because it will depend on income distribution and wage growth assumptions and so on over time. But I think that, broadly speaking, you would expect that the cost in the final year of the forward estimates would be a pretty good guide going forward, but you would expect that to move, and not necessarily in a way that was straightforward to calculate.

Senator STORER: The first round of the income tax plan is estimated to cost $13.4 billion over the period of the forward estimates. Is this cost offset by savings, or is it directly—

Ms Mrakovcic : It forms part of the entire budget, which would have set out a whole range of policy decisions on both the revenue and the expenditure side. I think that there are some reconciliation tables which essentially set out the contributions that changes in policy make.

Senator STORER: In information provided to the Senate on the revenue impact of the tax plan, Treasury says estimates are 'subject to heightened uncertainty'. What are those heightened uncertainties?

Ms Mrakovcic : I think it's just what I set out previously to Senator Ketter—that, essentially, we undertake the revenue estimates and we do costings on the best information that we have, but they are built around projections going out into the medium term, and the sense that we want to give when we provide this information is that the further out that you go from the current year the more that those uncertainties will tend to manifest themselves in terms of a greater degree of confidence interval, if you like, around your central estimate. So we have the statement in the budget documents which actually talks about the confidence interval around our revenue estimates and how that actually widens over time. But, Mr Ewing, did you want to make any observations on this?

Mr Ewing : It might just help to briefly explain the way that we do the modelling on this, which is that we start from a data file, which is the taxpayer population in 2015-16, and then have to make a range of assumptions about growth of different types of income and changes in the population to get us to 2024-25. So we have to make a lot of assumptions on top of our base data, and, as Ms Mrakovcic has set out, as we move forward in time, just the uncertainty around what the distribution of the population will be, distribution of wages, how much dividends will grow compared to interest income—all those sorts of factors—you get a growing uncertainty. So over time our estimates are subject to this heightened uncertainty.

Senator STORER: You spoke about out to 2024 being the period. Are they reflected year by year to that period, or is it in general?

Ms Mrakovcic : Sorry, I don't think we understand the question?

Senator STORER: I was just trying to understand. You run these figures out to 2024.

Mr Ewing : When we are looking at the medium term in the budget, we actually go out to 2028-29, which is the current end of the medium term. In a mechanical sense, models work year by year, but that doesn't necessarily mean that each individual year is a useful thing to present.

Senator STORER: In the statement to the estimates, Mr Fraser noted:

… there are large error bands around estimates made three or more years ahead.

Is there a margin of error that he's talking about in terms of that?

Ms Mrakovcic : I think he was probably referring to the statement 8 that we put into the budget, which basically talks about what I think you've heard me reference as a 'confidence interval' around our estimates. It simply flags that the further out you go the more that confidence interval, if you like, around your estimate widens—or your uncertainty band around your central estimate, you could anticipate, would widen.

Senator STORER: What would be the value then, when it's widening in the future out beyond the three or more years? What confidence could you take in the estimates of projections of that beyond 2022-23 to 2028-29? Are they worthwhile?

Ms Mrakovcic : We certainly think that doing modelling around medium and longer term—and we do it in a number of different projects or issues; for example, the Intergenerational report is produced and that goes out a very long way, and obviously we do the forward estimates, and we have done the forward estimates for pretty much all of my working life in Treasury. I'm pretty sure that that's the case. And more recently we have also supplemented that with analysis in the medium term. We think that it's important to have an understanding of policies and the implications of policies over the medium to longer term. I don't think we would advocate a scenario where governments would take a decision without referencing or understanding the medium to longer term implications of that, particularly if those policies were to continue outside the forward estimates.

But, by the same token, we have to be cautious and conservative in terms of acknowledging that, because of the way that we put forecasts together, they're built from the last year that we actually have historical data for, and from that point we are putting in our best assumptions. Over time, the aggregation of those assumptions means that it inevitably results in a greater degree of uncertainty around those estimates. We still think that they provide a useful guide, but we think it's important that people understand that they come with caveats and cautionary notes around their use, if that makes sense. I think it's a useful guide, but I guess that I worry, particularly where they can be used inappropriately on pinpoint accuracy, or to somehow or other design minutiae changes in policy under the assumption that five or six or seven years out you can change a policy design detail and know to $10 million or $20 million what the impact is going to be. Does that make sense?

Senator STORER: Yes. To that point, how prudent is it therefore to be locking in measures that would cost tens of billions of dollars, such as this stage 2 and stage 3, when we're talking about uncertainty of the economic environment beyond the horizon such as you're talking about? It doesn't seem to me to be very prudent to be doing so.

Ms Mrakovcic : Obviously these are decisions for government and parliament. The observation that I would make is that the government has articulated clearly that it has set a speed limit on revenues of 23.9 per cent of GDP. That number manifests itself or hits—I think that it in the last year of the forward estimates it is just shy of that figure, and that's after you take the personal tax plan into account. Not only that but if you then look at the medium turn, I think that even there, if you look at the unconstrained revenue situation, it butts against that speed limit again. Sorry, that was a very long way of saying that I think—given that the government has indicated that it has got that speed limit of 23.9, and given that in this budget there is the expectation of an improved fiscal outlook, and that starts as early as the immediate years—that that would give you some degree of confidence.

Senator STORER: The 23.9 per cent was not modelled by Treasury, correct? The figure from 23.9 was—

Ms Mrakovcic : The actual—

Senator STORER: The derivation of that.

Ms Mrakovcic : Its use as a speed limit?

Senator STORER: Yes.

Ms Mrakovcic : That's government policy.

Senator STORER: Yes. There was no process undertaken within the department to provide that to government. That was provided by government to the department, correct?

Ms Mrakovcic : I can't comment on that. All I can say is that it's government policy.

Senator STORER: Okay. Are there other tax initiatives legislated where substantial elements of the package are not to take effect more than six years in advance?

Ms Mrakovcic : The enterprise tax plan laid out a series of changes to the rates and thresholds over a long period of time.

Mr Ewing : Although it is perhaps only in part a tax measure, I recall that there was a superannuation guarantee rate increase schedule set out in the 2010 budget that extended well beyond the forward estimates as well.

Senator STORER: Should tax cuts be funded by offsetting structural savings to the budget rather than relying on cyclical improvements in the future, such as what we're relying upon here to provide these tax cuts, these projections that are, from present levels, higher in terms of wage growth and overall growth, and so they're relying on a cyclical upturn in the future to fund?

Ms Mrakovcic : Those issues are a matter for government. The only observation I'd make is that in this budget there was certainly an improvement in the overall fiscal outlook. There was an expectation that net debt would peak in 2017-18. Certainly, the fiscal aggregates represented an improvement over previous budget documents, and those aggregates did already themselves take into account the personal tax plan and the enterprise tax plan. On the estimates that are laid out, the fiscal position does move into balance and then into surplus, and remains over the medium term. That does represent our best sense of where the numbers are moving and where the economy is moving, and the application of the government's policy of a 23.9 per cent speed limit.

Senator STORER: Is it correct that the first stage costs are stated to be $16 billion for step 1, but, of the total $102 billion—that's step 1 and step 2 together—the vast majority are for changes occurring 1 July 2022 onwards? That's $41 and $36 billion, so 70 per cent—the lion's share?

Ms Mrakovcic : That's correct. The $102 billion refers to the measures that either commence on 1 July 2018 or 1 July 2022.

Senator STORER: Right now, in June of 2018, the proposal has significant—over half of its cost is due 1 July 2022 onwards, which is beyond the forward estimates?

Ms Mrakovcic : That's correct.

Senator KETTER: There was an article in The Australian on 30 May which dealt with Treasury costings of analysis of the government's tax plan compared to an alternative tax plan. Minister Cormann tabled, during the course of estimates, a document which had the wage rates which were underpinning that analysis. Are you familiar with what I'm talking about, Ms Mrakovcic?

Ms Mrakovcic : I am familiar with it. I do recall it being raised at estimates last week.

Senator KETTER: It's something you can take on notice for me, but could you confirm for me that of the 30 annual wage rates referred to in that document, apart from four or five, all are in the top 10 per cent of income earners?

Ms Mrakovcic : We'll take that question on notice.

Senator KETTER: Thank you.

CHAIR: I have a question about bunching, which is the phenomenon of taxpayers gathering behind a particular threshold. Do you have any data on whether that occurs?

Ms Mrakovcic : We have looked at the issue, and it certainly is a phenomena, that there is bunching when you have that change in the marginal tax rate. I'm afraid that we'd have to take the question on notice in terms of providing you with more detail on that.

CHAIR: That's all right.

Ms Mrakovcic : I can certainly—

CHAIR: It is a phenomenon that is recognisable in the Australian tax system?

Ms Mrakovcic : It is certainly a phenomenon.

CHAIR: And a flatter tax system would obviously prevent that from occurring because there aren't so many thresholds under which to bunch?

Ms Mrakovcic : Certainly to the extent that there are fewer thresholds to bunch around. You would expect less of that activity.

CHAIR: I would really appreciate some data on that if you have some.

Ms Mrakovcic : We're happy to take that on notice.

CHAIR: Terrific.

Senator KETTER: To help you with your question on notice on the matter of the millionaires that we talked about earlier: it was Grattan, and they've indicated that it's the AfrAsia Bank Global wealth migration review which was released recently. It looked at the migration of millionaires around the world. Ten thousand millionaires moved to Australia last year; almost none left. This is the highest net migration of millionaires to any country last year in absolute terms, let alone relative terms. I'm interested in whether you can shed some light.

CHAIR: I think that that information also had a country-by-country breakdown as to where the millionaires came from, which might be helpful in any analysis you do. I remind you and all witnesses that questions on notice are due by close of business next Wednesday, 13 June. Thank you very much for appearing before the committee today. That concludes today's hearing. On behalf of the committee, I'd like to thank all of those who have made submissions and sent representatives here today for their cooperation in this inquiry.

Committee adjourned at 17 : 12