Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Economics Legislation Committee
10/04/2019
Estimates
TREASURY PORTFOLIO
Australian Taxation Office

Australian Taxation Office

[14:15]

Senator KETTER: I return to my questions in relation to the super amnesty. The amnesty period was to extend for a 25-year period. What are the average liabilities per year declared over that 25-year period?

Mr O'Halloran : I'm not sure I have that either with me or even available. Could I just check I understand: you're asking how far back people disclosed?

Senator KETTER: What's the distribution over that 25-year period?

Mr O'Halloran : I don't have that and I'm not sure that analysis is even readily available. From some early data that I saw a while ago, it did appear that the pattern—I can't distinguish whether it was with an amnesty in mind or through the normal SGC process—most of it was clustered around the last five years. As for whether I have an analysis of that, I'm genuinely not sure. When I saw some cluster analysis on some pretty high-level data, it seemed to be that most of the disclosures were within that four or five-year period—as in, retrospective.

Senator KETTER: Would you have a view as to what the disclosures represent in terms of what would be the reasonable annual estimate of noncompliance?

Mr O'Halloran : The pattern or perhaps the amounts?

Senator KETTER: What you're familiar with is the most recent five-year period. If you look at each of those years, how do they compare with what might reasonably be expected to be the level of noncompliance?

Mr O'Halloran : One of my difficulties is by their very nature the hundreds of pieces of data that come in require a disclosure by an employer for each quarter for each employee. There are a lot of quarters. Other than that the numbers are high and the amount of payment is actually higher, because people have come forward and paid at the same time, I'd have to take that on notice, to be quite candid. I'd only be speculating on not much more data than what I've outlined here this afternoon.

Senator KETTER: Yes, could you do that, thank you. What checking does the ATO undertake, then, to assess whether the declared amount is actually correct?

Mr O'Halloran : There are probably two things. Normally through the SGC statement process it does come relatively complete in the sense of its specification of the employee, the periods et cetera, because otherwise it can't be processed by the nature of any payment that's made; we don't know whom to disburse it to. You are also then getting into a situation where the whole design behind the SGC statement process is for people to bring forward so that there's not a disproportionate allocation of resources on what should have been complied with in the first instance. There's foundational checking of that data. Whether it's referred for further work varies a little bit. I'd have to be quite candid: in a sense, it looks complete, it matches with some of the algorithms and there is manual checking of the documentation, the calculations and the manner in which it is presented. Sometimes that needs correction or some follow-up. That's the general essence. What I'm referring to there is not the casework component—as I touched on before, some 18,000 cases per year. That's probably the best response I can give without going into a lot of procedural detail that is probably not the essence of your question. We certainly do checks on the history of the people concerned, and those sorts of things, to see if there is anything that looks disproportionate, in line with our practice statement. Other than that, that is probably the way I would describe it for today.

Senator KETTER: How would you pick up that they have not declared the full amount they should have paid?

Mr O'Halloran : The short answer is that, like many of these things, we make some risk judgements in relation to that. Since this committee, and others, have raised questions around not relying on employee notifications, at least on paper, we are now running at about 45 per cent of all our casework being initiated as a result of ATO analysis, which includes a whole range of sources, which does give us a profile on the likelihood of non-compliance and the history of the employer concerned.

Senator KETTER: Who calculates the relevant SG charge amount?

Mr O'Halloran : It's part of the submission process. There are also automatic checks and balances on the calculations on the forms as presented.

Senator KETTER: Can you confirm that both the nominal interest charge and the general interest charge have been applied to amounts declared under the amnesty?

Mr O'Halloran : Yes.

Senator KETTER: And what's the value of each, in addition to the SG shortfalls declared?

Mr O'Halloran : I'd have to take that on notice.

Senator KETTER: What is the value of the part 7 penalties that have been waived under the amnesty? And I note you made the point earlier that they haven't been waived.

Mr O'Halloran : Yes, I won't go back over that. I would have to take that on notice. I have a figure in mind, but I can't be certain that that is specifically quarantined for part 7.

Senator KETTER: What's the value of liabilities that have actually been paid to the ATO under the amnesty and how many employees is that in respect of?

Mr O'Halloran : I thought I mentioned that before, but I am certainly happy to confirm out of session as well. We estimated approximately 5,000 employees. There has been $86 million paid to the ATO directly, there is $61 million under active payment plans—in other words, people are contributing according to a payment plan weekly, fortnightly or monthly, according to the terms and conditions—and there was $14 million paid directly into employees' funds. The sum of that is about—

Senator KETTER: What is the value of the actual payments that have been made to the super accounts of employees?

Mr O'Halloran : $14 million.

Senator KETTER: That has now been paid?

Mr O'Halloran : Am I missing the point? To employees?

Senator KETTER: Yes.

Mr O'Halloran : Sorry, I'll rephrase that in case I have missed it. The $86 million that has come into the ATO has been paid to the employees.

Senator KETTER: It has all been paid into the super accounts of the employees?

Mr O'Halloran : Yes, according to the identification of the employees that comes in. The $61 million is progressively paid as the terms and conditions are executed, over the weeks or months or whatever the individual payment arrangements are. As that comes in, that is then prioritised. The Super Guarantee is a high order disbursement point, so the money goes out to employees—into their fund. The $14 million—which really happened outside the ATO, but the ATO were notified—went straight into an employees' fund from the employer. Does that clarify that, Senator?

Senator KETTER: Yes. Thank you. Given that there has been no legislation passed, obviously employers won't be able to claim a tax deduction on the historical underpayments. How are you going to ensure that they don't?

Mr O'Halloran : As we move into tax time, obviously that will be the sorts of checks and balances we will be checking as we continue to sort this data once we are confident over the next period that the amnesty is not proceeding. But that would be a normal part of our risk because it is such a clean-cut issue, if I may say, to make sure that for identified employers—who, of course, we can identify from the material that has come forward—we would be able to set up a check. I'm not sure whether it will be automatic or manual at this stage, but it is in the order of 15,000 employers, which, on a relative scale, is a moderate size but not an unmanageable population.

Senator KETTER: Can you tell us: what evidence is there that the amnesty has resulted in a higher level of SG liabilities being raised and successfully paid to employees compared to normal compliance activity in previous years?

Mr O ' Halloran : Firstly, the only evidence I can give at the moment is that the incidence level seems to be somewhat higher. In other words, a higher number of employers have come forward. I think I mentioned 10 to 15 per cent. We haven't, as yet, done a comparison in any detail with previous normal SG submissions, other than on a numerical count. As I say, this is quite intensive, given that it's reported per period, per employee. I suppose that's the only early indicator: there's been a higher level of awareness as well as participation, if you like. I don't think we'd have much more at this stage. Certainly, as I mentioned earlier, anecdotally if not in practice, the profile of the acceptance or non-acceptance of not paying superannuation guarantee to their employees is changing in tone. In the next couple of weeks we will start to use some of the new enforcement tools. One, which we will use sensibly, progressively, within the law and within tone, is the ability for us to notify or contact employers where we reasonably believe—I think that's what the law says—or have a reason to believe that they haven't been paid their SG. It is an important feature because, until that announcement, we were limited by only being able to communicate with employees who had made a complaint—an employee notification to the ATO. We will be measuring the effectiveness of those. I'd like it sooner rather than later.

Obviously, we would like to think that, as people and employees are now aware—and I apologise if I'm going over old ground—and I put great store in visibility, and people can see their SG payments through ATO online services—and I'll be tracking it very closely—our employment notifications will, ideally, reduce over time. There are the 20,000 or 25,000 we get a year, which is employees who believe they were not paid super guarantee. That's a feature that will change aspects. The second one is the continued activity and better data as well, from our point of view. Our proactive or ATO initiated cases will have a much stronger profile from the macro data from the funds, from our own analysis and obviously from some of our casework. We are trying different models to—to be honest—be more assured that there are adjustments that can and should be done and that we can apply in an appropriate framework to collect the SG.

Senator KETTER: Will employers who have come forward under the amnesty be flagged and subject to future compliance activity or audits?

Mr O ' Halloran : The law never existed. As with any super guarantee payment—and I touched on this before—we don't see a distinction. Perhaps I could put to you the question which is a feature. If people have come forward under a super guarantee charge process in the normal course of events, that is a feature. If they've come up in audit before, one would need to be mindful. Dare I say: once is an accident, coming forward two or three times is a behaviour that may, in fact, impact if we are auditing them as a consequence of other things. Is it a major targeting issue—the fact that somebody has come forward within the current law to actually make arrears, if you like? It's a consideration, but not in the terms that you've described it. It would be part of the profile of an employer, similar to other issues. Sometimes it's beneficial; sometimes it might indicate that what they said was an accident wasn't an accident and is actually a course of conduct that, when audited, one would take into account. Perhaps there hasn't been good compliance history or there has been misuse of the opportunity to learn from mistakes in terms of late payment or underpayment.

Senator KETTER: If the underpayments have occurred over a significant number of years, wouldn't that raise alarm bells?

Mr O ' Halloran : Sorry, that's my point. It's one of the determining factors. It is one of a range of factors. Particularly in ATO initiated cases that come forward, that's the sort of background information that we would normally get. It's not in the sense of, if I may, coming up through the amnesty; it's part of the history of the interactions that the person has had with the ATO, when they come forward with the superannuation guarantee lodgement.

Senator KETTER: I'm talking the cohort of about 19,000 employers that you were talking about. Is there going to be an analysis of each of those declarations, to identify if there is a need for a follow-up—an audit, or some sort of compliance activity?

Mr O'Halloran : The short answer is no, not just because of that trigger—in other words, just because they've come forward. And it has always been open to people to come forward. That might be a good thing. That 19,000, of course, is the full number of people who have come forward under the existing SG charge process or arrangements. But if they come up through other processes then that would become part of a history, as opposed to targeting them because they've come forward, because, if it's been a full and frank disclosure and if it's been paid et cetera, then, in a sense, that's one of the reasons, including the nominal interest, that there is an ability for employers who discover, or realise, or choose to come forward, under the current law. Have I misunderstood your point, Senator?

Senator KETTER: No, that's okay. What's the ATO's policy going forward as to part 7 penalties? Is there a minimum 50 per cent or 100 per cent penalty—

Mr O'Halloran : No.

Senator KETTER: or is each case assessed on its merits?

Mr O'Halloran : Each case is assessed on its own. The practice statement, as it currently stands, outlines quite clearly that there are remission considerations—behaviour, characteristic et cetera. So, with reference to what had been speculated on, at this stage there's no change in the ability, other than administratively, for the ATO to consider each circumstance and balance a range of features, as opposed to a floor—or a ceiling; I'm not sure which way it is!—of 50 per cent. And I think, across all heads of revenue, it's always been a position. And it is, obviously, a matter of differing views at different times. But certainly the individual circumstances and so forth should always be taken into account. As I mentioned earlier, we have in fact tried to—through better information, rather than subjective judgement. We will see things like: for the first time, somebody hasn't paid an employee or group of employees SG for the last four periods, and we'll know it within 10 days or something of when the payment should've been made. So I think: once is an accident; two or three times becomes a pattern of conscious behaviour. That's where I think we will in fact be able to get a fair and balanced but also empathetic penalties regime.

Senator KETTER: My final question on the SG amnesty issue is this. In the past 12 months, how many 200 per cent part 7 penalties have been applied to non-compliant employers?

Mr O'Halloran : I know the first answer will be none, in terms of 200 per cent. If you can just bear with me, we have in fact—I'm sorry—

Senator KETTER: If the answer is none, that's the answer. That's what I'm—

Mr O'Halloran : But there has been an increase in the over 50 per cent and the over 100 per cent.

Senator KETTER: Well, if you've got those figures, I'm happy to have those.

Mr O'Halloran : I'll have to take it on notice.

Senator KETTER: All right. I know we have some further questions in relation to the ACNC, but, before we do that, I want to come back to that issue of net overseas migration, and perhaps, Ms Mrakovcic, you might be able to help us with those. There was a letter that I'd sent as to the estimates of net overseas migration for the 2019-20 budget forward estimates, and I think Senator McAllister asked some questions in relation to this issue. As to Treasury's response to my letter from last week—and there was a subsequent response to that—I just wanted to firstly confirm that, compared to the last budget, net overseas migration is roughly 40,000 higher per year.

Ms Mrakovcic : I might ask Mr Brine to answer those questions.

Mr Brine : Sorry—I have the figures for the permanent migration program here, but I don't have the NOM figures with me; they are in BP3, but I can bring that document along. Macroeconomic Group prepares the NOM figures and is responsible for the composition and the level of those. We sort of translate the NOM figures into revenue forecasts. There was a lot of concentration in this budget update around changes to the permanent Migration Program and how they feed through to revenue, and I can certainly help with any questions in that space. But NOM, as you know, is composed of the permanent Migration Program, changes in temporary visa levels and also arrivals and departures of Australians, and that's very much in the Macroeconomic Group's space.

Senator KETTER: Did you say you have another file that has information in respect of this matter?

Mr Brine : In relation to NOM?

Senator KETTER: Yes.

Mr Brine : Another file?

Senator KETTER: Your first comment was that—

Mr Brine : If there are questions about the permanent Migration Program changes that were announced in the budget context and how they feed through to revenue, I can certainly help with those. But, on the NOM figures, my understanding is they were a little bit below 260,000. They increased to a little bit above 270,000. But I just don't have the comparison to previous budgets for that overall figure here.

Senator KETTER: Okay.

Senator McALLISTER: It's a very significant comparison. I don't have my documentation with me either, but I did ask questions about this last time. It's essentially somewhere between 40,000 and 46,000 higher than the figures that were used in MYEFO, I think. So it's a very significant change.

Mr Brine : My vague recollection is that change—

Senator McALLISTER: My apologies—it was in comparison to the last budget. Looking from the last budget to this budget you get around 40,000 additional people.

Mr Brine : My vague recollection is that there has been a significant increase in the number of students. But, again, this is the Macroeconomic Group's territory. I offer that to be helpful, but I'm certainly not an expert in the NOM figures.

Senator McALLISTER: On the assumptions you're making in terms of revenue, there's something like an 80,000-person increase in migration above the permanent cap. If you think the permanent cap is going to be 160,000 and Budget Paper No. 3 tells us that in 2019 there's going to be a 271,000-person increase in net overseas migration it's about 80,000 people who are coming in by means other than the permanent program. What proportion of those 80,000 people do you assume have work rights and, therefore, flow through into revenue from an income tax perspective?

Mr Brine : I don't know that exact proportion. I do know that there were around 1.7 million migrants living on temporary visas as at December 2018, and that 1.7 million included New Zealanders living in Australia, students, working holiday-makers and temporary skilled workers. But I just don't know the split of those or what's driving that change in the 1.7 million over time.

Senator O'NEILL: Can I ask a clarifying question about information we received from the Fair Work Ombudsman the other evening. This is from memory, but I'm pretty sure 758,000 was the number they were talking about.

Mr Brine : For temporary visa holders with work rights?

Senator O'NEILL: Yes. So if that number of 758,000 is correct, can you explain the other million?

Mr Brine : That would be New Zealanders living in Australia, students and working holiday-makers.

Senator O'NEILL: How many?

Mr Brine : I don't have those numbers available, sorry. I have a total, but I don't have the decomposed numbers. Again, these are questions that are really the Macroeconomic Group's responsibility.

Senator O'NEILL: But the revenue numbers are your responsibility.

Mr Brine : They give us the NOM figures. Yes, that's right.

Senator O'NEILL: So that's the numbers of people who were there, but the assumptions that they would then make about how many of those people with rights would actually activate those rights must affect your revenue number? The Fair Work Ombudsman were quite clear that they didn't have a sense of how many people were working. They did with students, but with the other groups they didn't know. So I'm trying to figure out how you can make up these numbers and stand here and support them if you can't answer questions which are interrogating the value of the data that you're attributing to another entity that can't answer the question here.

Mr Brine : There are two products we provide in the Tax Analysis Division. One is the costings of policy changes and one is the revenue forecasts. In the costings of policy changes, we do a fairly detailed disaggregation of visa classes. We draw on our microsimulation models, where we can see the amount of tax being paid by people on different visa classes, so we can do a pretty granular assessment of how change in planning levels for different visa class levels will impact the budget over time. For the revenue forecasts, we use a base-plus-growth model, where we take the last observed outcome for each revenue head and we grow it forward according to different parameters. So we might grow individuals' income tax in line with compensation of employees, or we might grow other income in line with gross mixed income, or we might grow company tax and certain company tax income streams and deductions in line with gross operating surplus.

Those components come to us from MEG, and they have built their NOM estimates into those components—into those elements of the macroeconomic forecasts. So I don't need to get their NOM numbers. I rely on them to have fed their NOM numbers through into their estimates of compensation of employees, GMI and GOS. They give me those numbers and then I grow the base of the different revenue heads using the appropriate parameter.

Senator O'NEILL: I understand what you're saying to me. I still don't necessarily have confidence that that's of sufficient granularity given the differentiation about what's going on in each of those cohorts was unknown to the Fair Work Ombudsman. Who's got the clarity of view of the—

Mr Brine : Macroeconomic Group would be the people who make the assessment of how the changes in NOM will feed through into things like COE, GOS and GMI.

Senator KETTER: In Budget Paper No. 3, for net overseas migration there's a table, A.2, on page 92. Does that assist you with—

Mr Brine : That's the figure I mentioned before. We're slightly below 260,000 in 2018, the figure increases to slightly above 270,000 in 2020, and then it comes back a little bit. That's definitely the total NOM figure. The componentry sitting underneath that total I can help with in some respects—I can explain how those NOM figures feed through into our revenue forecasts—but exactly what is driving that change over time and why the NOM figures might be different to previous years are really questions for our Macroeconomic Group colleagues, who rely heavily on the immigration department.

Senator KETTER: There is a roughly 80,000 person increase in migration above the permanent cap. Are you able to tell us what proportion have work rights?

Mr Brine : No. As I think I mentioned before, I don't have information on what proportion of those—

Senator KETTER: But isn't that information necessary for you to look at the revenue implications?

Mr Brine : No. As I mentioned before, we do costings on a fairly detailed level. We do costings for things like government policy changes, such as change in the permanent migration program. We would do detailed analysis of visa classes, looking at the amount of tax paid, the amount of income tax and indirect taxes paid by people coming in on different visa classes. We can estimate pretty accurately how changes in individual visa classes will result in changes in revenue. For the revenue forecasts, we use a different model. We use a base-plus-growth model. We take the last observed tax outcome, which at the moment would be the 2017-18 year. We then grow that in line with the appropriate economic parameters. The appropriate parameter, say, for individuals' income tax would be compensation of employees. So the growth in the amount of the economy that we think is going to end up in employees' wages is the parameter that we use for individual income taxes. That parameter comes from our colleagues in Macroeconomic Group, and in producing that parameter they will take into account the NOM forecasts. That occurs in another part of Treasury and we get the CoE forecasts—compensation of employees, gross mixed income, GOS. Other parts of the economy feed through to other parts of the revenue forecasts. So we don't need to know the exact NOM numbers for the revenue forecasting, which, as I mentioned, is different to the costing.

Senator KETTER: So you wouldn't be able to help us with the number of people that have work visas compared to student visas?

Mr Brine : That's correct. That would be something that Macroeconomic Group would take into account, but they wouldn't necessarily give us visibility of that.

Senator KETTER: Chair, I would like to explore that issue further. I'm not sure whether or not we can look at having Macro come back, if Macro is the group that is best placed.

CHAIR: I'm very reluctant to call back a group for a third time.

Mr Brine : Could I help assist. While Macroeconomic Group is responsible within Treasury, of course the immigration department are the experts in this and they would be feeding information to Macroeconomic Group. If there were someone to provide granular information about net overseas migration, it would be the immigration department.

CHAIR: And Home Affairs have come and gone too.

Mr Brine : Sorry, Home Affairs.

Senator O'NEILL: They're in witness protection. We can't get them up in the Senate. They'd have to answer serious questions about Minister Dutton.

CHAIR: We did have them for two days earlier in the week.

Senator KETTER: I've got some other questions in relation to birth rate assumptions that have been used to inform population growth forecasts in the budget. Is that something you're able to assist with?

Mr Brine : Again, that would be Macroeconomic Group.

CHAIR: Senator Ketter, I'm conscious that Senator Siewert has more questions for ACNC.