Home | Databases | WorldLII | Search | Feedback James Cook University Law Review

Woellner, Robin --- "Updating the Tax Dispute Burden of Proof Provisions - in Light of the Expansion of ATO Pre-Filling" [2024] JCULawRw 5; (2024) 30 James Cook University Law Review 69

Updating the Tax Dispute Burden of Proof Provisions – In Light of the Expansion of ATO Pre-filling

Robin Woellner^[1]*

Abstract

Pre-filling of personal tax returns by the ATO has been a hot topic in Australia in recent times. In brief, pre-filling involves the creation of tax returns fully or partially completed by the revenue authority (in this case the Australian Taxation Office (‘ATO’)) based on information supplied to it by third parties such as banks, companies, employers, health funds and a wide range of other bodies holding information relevant to a person’s tax position. However, while much has been written on the topic of pre-filling,^[1] nothing appears to have been written about one of the key flow-on effects of pre-filling – the possible impact which pre-filling might have on the rationale for imposing the onus of proof in tax disputes on the taxpayer. This latter issue is the focus of this article.

I The Current Burden of Proof Provisions: Taxation Administration Act 1953 (Cth) ss 14ZZK and 14ZZO

The principles underlying the tax administrative provisions currently located in Tax Administration Act 1953 (Cth) – including the crucial decision to place the onus of proof in taxation disputes on the taxpayer – have remained largely unchanged for many years.

Thus, under ss 14ZZK(b)(i) and 14ZZO(b)(ii) of the Taxation Administration Act 1953 (Cth), so far as relevant:

In proceedings for review [by the AAT] of a reviewable objection decision:

(b) the appellant (i.e. the taxpayer) has the burden of proving [on the balance of probabilities]:

(i) if the taxation decision concerned is an assessment – that the assessment is excessive or otherwise incorrect and what the assessment should have been or

(ii) in any other case – that the taxation decision concerned should not have been made or should have been made differently’.

Paragraph 14ZZO(b)(i) and (ii) make analogous provision for appeals to a court of an ATO objection decision.^[2]

As Derrington J observed in FC of T v Ross:^[3]

... the rationale for the onus imposed by s 14ZZO) (b)(i) is that the facts relating to [a taxpayer’s] taxable income, including the work that they have undertaken and the transactions they have entered into, are peculiarly within the taxpayer’s knowledge, and they must be taken to know what their income is and how it was derived.^[4] Conversely, the Commissioner has limited knowledge of those circumstances. On that basis, there is no undue harshness in requiring a taxpayer, who has failed to lodge a return or whose return is not compliant with the taxation legislation, to bear the onus of establishing their true taxable income for each income year.^[5]

Over the years, the courts and tribunals have developed numerous principles or propositions relating to the burden of proof in tax disputes. Among the many propositions found in the case-law^[6] in relation to the tax burden of proof provisions, some key principles include:

• The ATO has no obligation to positively prove that the assessment is correct, or indeed to prove anything, and the ATO can rely upon any deficiency in the taxpayer’s case (the reverse does not apply). Generally, under s 167 ITAA36 (Cth), the taxpayer must positively prove the amount of their taxable income,^[7] and demonstrate that this amount is lower than the amount assessed – i.e. they must demonstrate that the assessment is excessive’.^[8]

• It is generally not sufficient for the taxpayer to simply point to flaws in the ATO’s case (for example, that certain amounts are not income) because this does not prove that the assessment is excessive.^[9]

• Challenging a default assessment made under s 167 ITAA 36, (for example, on an asset betterment basis) can impose particularly significant difficulties for a taxpayer,^[10] because the taxpayer must demonstrate that any unexplained accumulated wealth was derived from non-income sources. It has been said that this may be achieved by an accepted denial of any undisclosed source of income, providing acceptable evidence of how the taxpayer spends their time, and demonstrating a reasonable explanation for any appearance of the possession of assets ...’^[11]

As Derrington J acknowledged in Ross: ‘These principles can result in a situation where [a] default assessment can be assumed to be inaccurate in some respects but, in the absence of the taxpayer establishing what their actual taxable income was, it must nevertheless stand’.^[12] That is, the burden of proof can create situations where an assessment which is acknowledged by all parties to be wrong is nevertheless upheld by the decision-maker simply because the taxpayer is deemed to be unable to satisfy the burden of proof.

These burden of proof provisions are therefore extremely important because they impose a heavy probative burden on taxpayers – indeed, not infrequently, the allocation of the burden of proof will determine the ultimate outcome.^[13] As a result, the burden of proof can create situations where an assessment which is acknowledged by all parties to be wrong is nevertheless upheld by the decision-maker simply because the taxpayer is deemed to be unable to satisfy the burden of proof. It is therefore crucial to revisit the burden of proof provisions and their de facto operation regularly, in order ensure that despite the passage of time and other changes, they are still appropriate^[14] for the current and developing contexts.

When they were first introduced – long before computers were in use – the current burden of proof provisions were clearly logical and appropriate – for example, in the dim distant past, some taxpayer records were held by the ATO on paper cards, which meant that searching for specific information or cross-references on a particular taxpayer or taxpayers could be an onerous, tedious, time-consuming and potentially frustrating task.

Accordingly, in the past, where the taxpayer generally had more direct knowledge of (or at least comparatively easy access to) their tax details, but the ATO may often have had little or no (easy) access to such information, imposing the onus of proof on the taxpayer was arguably logical and reasonable.

However, times have changed dramatically, and, in the modern computerised era the roles of ATO and taxpayer have arguably been reversed. Taxpayers will always have some knowledge of their tax affairs, but the ATO now has instant access to massive amounts of digitised data and to sophisticated computerised search engines which can provide almost instantaneous and refined cross-referencing to various aspects of a taxpayer’s activities and activities of related persons. Most taxpayers will not have access to all of this data, so that the ATO may often now have more timely, accurate and detailed access to relevant tax-related data than will the taxpayer. It may therefore be an appropriate time to reconsider the relevance and appropriateness of the existing burden of proof provisions, in order to ensure that they remain fair and fit for purpose in the modern world.

II Key Aspects of ATO Pre-filling – An Overview

A The Process

As noted, ‘Pre-filling’ or ‘e-tax’ is the process whereby the ATO inserts details into a taxpayer’s putative tax return, based on information which the ATO receives from specified third parties such as employers, companies, financial institutions, private health funds, businesses, government agencies and various other bodies which hold information relating to a taxpayer’s affairs.^[15]

As at 2023, the ATO advised that information categories required^[16] included categories such as: ^[17]

• Employee share schemes.

• CGT property transfer.

• Personal Superannuation Contribution Deduction.

• PAYG payment summaries and STP income statements.

• Single Touch Payroll (STP).

• Individual and non-business.

• Employment termination payment.

• Australian annuities and superannuation income stream.

• Superannuation lump sum.

• Business and personal services income.

• Foreign employment.

• Government payments.

• Dividend and interest schedule.

• Dividend income.

• ATO interest.

• ATO interest – recurring data issues.

• Interest income.

• High-certainty interest data.

• Employee share schemes.

• Managed fund distributions.

• Partnership distributions.

• Foreign source investment income.

• Crypto currency disposal.

• Work-related car expenses.

• Share and unit disposals.

• Property transfers.

• Business transactions.

• Taxable payments.

• Contractor payments.

• Government grants.

• Net farm management deposits or repayments.

• Tax offsets.

• Medicare levy surcharge.

• Private health insurance (PHI) policy details.

• Early-stage innovation company.

• Income test related items.

• ATO data.

• Help and other income-contingent loans debts.

• Prior year amounts.

• PAYG instalments.

• Accumulative low rate cap.

• Income averaging for primary producers and special professionals.

• Overdue income tax returns.

• Personal superannuation contribution deductions.

• First home super saver scheme (FHSS).

• Prior year tax return details. and

• Deductions.^[18]

Once a return has been ‘pre-filled’ by the ATO, it is sent to the taxpayer, who is expected to check the pre-filled return and to notify the revenue authority whether the taxpayer accepts the return or seeks to make of any proposed changes.^[19]

Scandinavian countries pioneered pre-filling: Denmark in 1988 was the first country to introduce pre-filled tax returns, followed by Sweden (1995) and Norway (1998).^[20]

Australia was a relatively late adopter of pre-filling, which was first mooted for introduction in 1998, but was not formally introduced to Australia until 2006/2007, following ‘minor trials’ in 2004/2005.^[21]

Currently, pre-filling is limited to personal taxpayers’ returns, though the ATO has indicated that it aims to develop more comprehensive pre-filling in the future.

B Suggested Benefits of Pre-filling

The benefits of pre-filling are argued to include:^[22]

• Greater certainty for taxpayers and revenue authorities,^[23] with resultant lower compliance costs for relevant taxpayers^[24] (in terms of reduced taxpayers’ personal time in storing and locating records).

• Reduced administration costs for revenue authorities and greater accuracy in tax returns, with reduced need for ATO tax auditing and other post assessment activities, particularly as the amount of pre-fill data expands.

• Improved levels of taxpayer compliance.

• Ultimately, the possible removal of the need for taxpayers to lodge tax returns,^[25] generating reduced compliance costs for taxpayers, and – perhaps – an increased respect for the revenue authority if pre-filled returns are seen by taxpayers as a genuine, personalised, and simpler service.

• Helping individuals and businesses understand their tax obligations.

• Serving pre-filled returns on taxpayers may remind taxpayers of their annual tax obligations to complete (or check) annual returns carefully and may also generate greater community confidence in the integrity of the tax system. However, this may be a two-edged sword – see below.

• Protecting honest businesses from unfair competition.

• Simplifying the tax collection process and making it easier for revenue authorities to assess the level of voluntary compliance and identify ‘problem’ areas.

C Suggested Problems with Pre-filling

While pre-filling offers many potential benefits, it may also create some potential problems,^[26] including:^[27]

• Pre-filling imposes administrative costs on those third parties required to provide information to the government.

• It has been said^[28] that a comprehensive pre-fill system also imposes costs on the revenue authority,^[29] viz: it needs to invest substantial resources, effort and persistence over a considerable period to establish and maintain a comprehensive system of pre-filling tax returns.

 Pre-filling assumes that the revenue authority receives accurate, complete and timely information,^[30] and correctly analyses it using appropriate computer programmes. These are not always safe assumptions in light of recent experiences in Australia with national and international computer-related problems,^[31] including those noted by Brookes and Dery.^[32]

 While receiving an annual pre-filled tax return might remind taxpayers of their tax obligations (see above), it has been suggested that on the other hand, shifting primary responsibility for returns from taxpayers to the revenue authority may mean that because taxpayers no longer need to actively consider their tax affairs, this may result in a reduction in taxpayers’ sense of responsibility for their tax affairs (including the accuracy of ‘their’ returns and other tax obligations), and result in taxpayers actually paying less care and attention to their tax affairs.^[33]

• A 2008 UK parliamentary report found that pre-filling ‘introduced a feeling of concern among taxpayers’ and reduced their willingness to challenge the government’s calculations’.

• It has been argued that There may be an increased risk of errors with a pre-fill system.^[34]

D The Expanding Scope of Pre-filling

Pre-filling in Australia began (relatively) modestly,^[35] focussing largely on the ‘easy’ categories of income and while the categories below may seem expansive, they are not comprehensive: the list of required disclosure items currently concentrates on categories of personal taxpayers’ income.

However, as with other jurisdictions, the categories of information which must be forwarded to the ATO have expanded (and continue to expand) significantly over time. Thus, in 2023, there were some 47 categories (set out above^[36] of required disclosure, as set above, while the ATO continues to add requirements – for example, in light of the increase in people working from home, the ATO announced that it would be monitoring closely claims for work-related deductions,^[37] and that from 2023 it had been receiving new data sources on claims by rental investors, including investment loan data, data on landlord insurance policies, and income protection insurance (deductions) and payouts (income).^[38]

Indeed, current limitations in the scope of pre-filling are largely a function of the comparative youth of the ATO programme and the complexity of the Australian tax system, and the ATO has stated that it:

continues to expand its data-matching capability to ensure taxpayers don’t leave out income or inflate deductions this tax time. It pre-fills some categories of deductions, and has signalled that it will continue to expand its pre-filling programme^[39] – and that it is aiming to create a comprehensive programme in time.^[40]

Indeed, Assistant Commissioner Tim Loh stated in June 2023 that taxation:

... isn’t a game of Guess Who, as our sophisticated data-matching programs provide us with all the clues we need to track down taxpayers with incorrect information in their tax return. We will use this information to identify and educate taxpayers who have made incorrect claims in their return, with a longer-term plan to pre-fill as much information as possible in future years ...^[41]

It is not immediately clear whether the latter comment means that the ATO intends to continue limiting pre-fill to personal tax returns or is intending eventually to extend the pre-fill programme tother categories of taxpayer.^[42]

In terms of possible future developments which the ATO might seek to follow, a number of Scandinavian countries have adopted full pre-filling for personal taxpayers.^[43]

This issue is discussed briefly below.

E Challenging a Pre-fill Amount

The ATO has confirmed that taxpayers are able to challenge or alter pre-filled data by contacting the ATO,^[44] though the burden of proof will continue to rest on the taxpayer. The ATO has also indicated that while taxpayers can challenge a pre-filled amount, the ATO may nevertheless decide to reject the taxpayer’s figures and instead adopt the pre-fill amount if the ATO has ‘a high level of confidence in the data’ it has received from other sources, and may ‘in some instances contact the taxpayer to ask why they had changed the pre-filled data, and to advise the taxpayer that they will need to provide sufficient evidence to support the change, otherwise the ATO may adjust the return ‘to align with the third party data’.

In fact, the ATO advised that in 2023 it had adjusted over 400,000 individual tax returns before issuing tax assessments, using third party data that was available via pre-fill but was omitted from the taxpayer’s return, and had also adjusted around 180,000 individual taxpayer returns with discrepancies after a tax assessment had been issued, based on validating information using third party sources.^[45]

That is, in terms of Derrington J’s analysis, in this situation the ATO is inserting much of the data rather than the taxpayer, and where it has a high degree of ‘data certainty’ in relation to the information it receives, takes the view that its data is superior to the taxpayer’s data, so that a taxpayer who wants to challenge a pre-filled amount must accompany their request with ‘cogent evidence supporting the requested change’.^[46] This is an example of the shift towards reliance on ATO-generated data rather than taxpayer-generated data.

The ATO’s approach is understandable, but this ‘second-level’ ATO requirement for ‘cogent evidence’ requirement is an administrative addition, not found in the legislation.

In light of the ATO’s approach, it seems that there is now a de facto double burden of proof applying in tax matters: in addition to the legal burden of proof in ss 14ZZK and 14ZZO of the TAA, there is now – semble – an operational ATO assumption that its pre-filled data is accurate and that the taxpayer’s data is not.^[47] This assumption may not always be correct, as experience has identified a variety of errors in pre-filled data (including, for example, bank accounts established on behalf of a taxpayer but unknown to them).

In seeking to expand the scope of pre-filling in the future, the ATO has a range of models it could choose to follow, though each system is partly shaped by the nature of the particular country’s tax system, and Australia’s tax system has a number of features which would affect the shape of a comprehensive system^[48] (including a complex tax system – in contrast to the Nordic countries, which tend to have comparatively simple tax systems).^[49]

As noted, in terms of trends/models the ATO could consider applying, in 2008 the OECD Forum on Tax Administration: Taxpayer Services Subgroup^[50] reported that:

... in Denmark, Norway and Sweden, pre-filled tax return systems have become quite sophisticated, having evolved to the point where most personal income taxpayers were receiving a pre-filled return that was fully completed, ‘while for many other taxpayers little further action was required...’ (emphasis added).

... For each taxpayer, the return discloses known income,^[51] deduction amounts^[52] sources and other calculated amounts, and provides a preliminary calculation of tax payable.

At the next stage of the pre-fill process, the taxpayer examines the pre-filled return, adds relevant information and notifies the revenue authority of any disputed amounts, and whether the taxpayer accepts (or is deemed to accept) the pre-filled return.^[53]

As an aside, it is interesting to note that penalties seem to play only a very minor role in many overseas reporting systems: the Swedish system does not apply any specific penalties for non-reporting by third parties, while the Danish system imposes only a small daily penalty on non-reporting.^[54] By contrast, in Australia, failure to provide information when and in the form required could be a breach of s 8C of the Taxation Administration Act 1953 (Cth), generating potential exposure to substantial penalties.

III The Burden of Proof Provisions in the 21st Century

Whatever the situation may have been in the past, many modern taxpayers, facing the spectre of the ATO and its serried ranks of supercomputers are now at a significant disadvantage in attempting to defend themselves. Nevertheless, the burden of proof currently remains steadfastly fixed on the taxpayer.

While the ATO has not yet achieved full pre-filling for all taxpayers), as Assistant Commissioner Tim Loh has made clear^[55] the ATO is committed to expanding its pre-filling programme and is working towards that end.

In terms again of Derrington J’s analysis, (above), the more the balance shifts from reliance in tax returns on taxpayer-generated data to reliance on ATO-generated (pre-filled) data, the more this erodes pro tanto the original rationale (discussed above) for imposing the burden of proof on taxpayers.

As discussed above, in those early days it was logical to impose on taxpayers the burden of proving on the balance of probabilities that (generally) an assessment disputing the information in their tax return was excessive.

However, as noted, those early days have been replaced by the widespread use of increasingly powerful and sophisticated ATO computer systems, which have dramatically altered the balance. The ATO now receives vast and ever-increasing amounts of digital data, which it can process and cross-reference almost immediately through the use of increasingly sophisticated search engines,^[56] rather than having to deal with hard copy returns which may require re-keying or other attention.

In addition, it is important to note – again in terms of Derrington J’s analysis – that the data which the ATO now obtains, processes and uses to pre-fill tax returns is increasingly created not by taxpayers, but by an ever-expanding army of third parties.^[57] The result can be that under pre-filling, a taxpayer may not see some of ‘their’ tax return data (with which they may disagree) until they receive a pre-filled return from the ATO. That is, more and more data in personal tax returns is being created by and on behalf of the ATO^[58] rather than by taxpayers and their advisers.

Overall, in a context where more and more data in tax returns is being pre-filled by the ATO, it seems increasingly illogical to impose the burden of proof on taxpayers (who may be asked to deal with data of which they were unaware or with which they may disagree), and increasingly logical to transfer the burden to the ATO.

While pre-filling only applies at present in Australia to personal tax returns, it would be wise to be alert to and prepared for possible future expansion of pre-filling to other categories of taxpayer. Such an expansion might be difficult to achieve at present, but increasingly widespread use of computerisation for record-keeping and increasingly rapid developments in technology can be expected to enable extension of pre-filling to other taxpayer categories in the future.

In this context, it is useful to recall that only some 30 odd years ago, pre-filling of any type was not even being considered as a possibility in Australia, before rapid developments in technology made the unimagined possible and increasingly available. In the same way, further developments in technology and related fields can be expected over time to rapidly expand future pre-filling possibilities to other categories of taxpayer.

IV What Test Should Be Applied To Determine the Appropriate Allocation of the Burden of Proof?

There are a number of possible approaches which could be applied in order to determine who should bear the burden of proof in tax disputes, but the most logical approach would seem to be to impose the burden of proof on the entity providing the (bulk or a substantial part) of the data used in assessing taxpayers, that is to adopt an approach of contra proferentem).^[59]

As discussed above, this would suggest that since the ATO now provides most of the data, the burden of proof in relation to individual taxpayers (who are by far the largest group of taxpayers and are the source of over 90% of the data used in tax assessments) should now be shifted to the ATO.

If this were done, then applying the same approach as in TAA ss 14ZZK and 14ZZO would suggest that ideally the same outcome should apply to all categories of taxpayer. This approach is straightforward and clear: all taxpayers are subject to the same burden of proof – there are no ‘special’ rules for trustees, partners or other categories of taxpayer. By parity of reasoning, if the burden of proof is transferred from the taxpayer to the ATO, the same burden of proof should apply in all cases. That ‘rule’ has operated successfully for a great many years. It also appears to be the most logical approach – there seems no reason to change a successful system.

Naturally, the overriding requirement for any shift of the burden of proof would be that any proposed shift would be workable in practice.

It is important to note that once the allocation of the burden of proof has been determined, any necessary rewording of the relevant provisions in ss 14ZZK and 14ZZO to accommodate the changes would be relatively straightforward and simple.

V Possible Future Approaches

Depending on future developments, there are a number of possible scenarios which might arise. Three possible approaches are discussed below. though it is likely that opinions on the preferable system will vary.

A Status Quo – Pre-Filling Remains Limited to Individual Taxpayers

While this would be an option, it seems very unlikely in reality that the ATO would revert to the ‘old’ system – pre-filling is far too well established for the ATO to make a U-turn at this stage and go back to the pre-computer days. It therefore it seems far more likely that the existing principles will continue to apply to individual taxpayers, at least for a period.

Given that under pre-filling of individual returns, the ATO currently provides a substantial part of the tax return data, applying the reasoning above would suggest that the ATO should bear the burden of proof in relation to individual taxpayers. Applying the current approach in ss 14ZZK and 14ZZO, (above) this would then suggest that the burden of proof for all taxpayers should be transferred to the ATO.

This approach would have the advantage of continuity and familiarity and would therefore be likely to cause least disruption and uncertainty for both the ATO and taxpayer advisers.^[60]

There might be a question of whether the transfer of the burden should be delayed until the ATO is actually able to pre-fill non-individual data. If so, it might be a consolation to note that that while any delay in modernising the burden of proof would be unfortunate, the increasing use of and reliance on digital record-keeping by all sections of the population and the blinding acceleration in the speed and sophistication of technology development suggests that any such wait should not be over-long.

As always, it would be essential to monitor developments regularly and closely, in order to identify objectively when the ATO reached the ‘tipping point’ which indicates that the burden of proof in relation to non-individuals should also be transferred to the ATO.^[61]

B Introduce a ‘Split’/Bifurcated System

If adoption of a unitary model was not possible or acceptable, an alternative approach might be to introduce a ‘split’ or bifurcated system.

This would apply separate burdens of proof to different categories of taxpayer, with, for example, the ATO bearing the onus of proof in relation to issues involving individual taxpayers, and taxpayers bearing the burden of proof in other cases.

Such a bifurcated/split approach would be possible, but it would not be without potential problems. It might, for example, create issues of demarcation between individual and other taxpayers (for example, when determining the category into which a particular taxpayer might fall – which could be exacerbated if a multiple split system applied separate burdens for each sub-category of taxpayer). It would also ignore the implications discussed above from ongoing developments in technology, which, in time, will inevitably solve difficulties currently facing the ATO in accessing non-individual data.

It would of course be essential to regularly monitor developments to identify when the ATO reached the appropriate level of data control to enable it to pre-fill significant amounts of non-individual data, making it appropriate to transfer the burden of proof to the ATO.

C The Preferable Approach

It would seem preferable to apply (or at least prepare to apply) a single burden of proof test such as that in ss 14ZZK and 14ZZO that will accommodate future technological developments as they occur. There could be an issue as to whether the burden of proof should therefore be transferred to the ATO now, to recognise the trend towards the increasing impact of technology, or deferred until the ATO actually reaches its tipping point, and whether a ‘split’ system, with all its problems, might be introduced as an interim measure?

While a single test covering all taxpayers and imposing the burden of proof for all categories of taxpayer on the ATO would clearly be the preferable option in a perfect world, this might not be achievable immediately, as the ATO apparently does not at present have the capacity to pre-fill non-individual returns.

However, it seems inevitable that the ATO will develop this ability in the foreseeable future, and it will be important to ensure that progress within the ATO is monitored regularly so that when its tipping point is reached, the government and ATO can move quickly to update its administrative processes and shift the burden of proof.

VI Pre-Filling Expands Beyond Individual Personal Taxpayers

Accepting that technology will continue to advance and refine the ATO’s data-gathering capabilities, it would – as always – be essential for the government and interested bodies to regularly monitor developments, to identify if and when the ATO’s data gathering abilities reach the ‘tipping point’ where the ATO is able to access significant amounts of data in relation to entities other than individuals.^[62]

When this future point is reached, and the ATO is able to extend pre-filling to provide significant data in relation to non-individual taxpayers, the appropriate allocation of the burden of proof the situation would be quite straightforward: all taxpayers should then operate under the ATO pre-filling system, with the ATO bearing the onus of proof in all cases.

This would essentially mirror the present situation, which has operated reasonably successfully for many decades and – significantly, would also be completely consistent with the rationale (discussed above) for the original allocation of the burden of proof.

Once it is decided to transfer the burden of proof, the legislative amendments to ss 14ZZK and 14ZZO needed to achieve the transfer would require minimal drafting adjustments. For example, the existing s 14ZZK(b) could be amended to read along the lines that:

In proceedings for review by the AAT of a reviewable objection decision:

(b) the ATO has the burden of proving on the balance of probabilities in relation to all matters in dispute that the assessment concerned was validly made and accurately determines the taxpayer’s tax liabilities.

An analogous amendment could be made to s 14ZZO.

A Could It All Be Much Ado About Nothing?

The possible impact of any such changes on averments by the ATO in prosecutions for prescribed taxation offences and the Evidence Act 1995 (Cth) is intriguing.^[63] Under TAA s 8ZL(1), in a prosecution for a prescribed taxation offence (as defined), a statement or averment contained in the information submitted by the ATO is prima facie evidence of the matter so stated or averred.

Under that provision, if the onus of proof was shifted to the ATO, as has been suggested above, is it possible that even in Part IVC cases the Commissioner could simply rely on a certificate as prima facie evidence of the particulars stated, thus effectively pushing the transferred onus back onto the taxpayer? Section 69 of the Evidence Act 1995 (Cth) (admissible hearsay in business records) has a (somewhat) similar effect.

The inter-relationship and effect of these provisions would need to be considered in any amending legislation.

VII Conclusion

As discussed above, with the impact of increasingly computerised record-keeping and the ongoing development and refinement of increasingly sophisticated computerisation and other technical advances, the process of preparing and analysing tax returns has changed dramatically over (relatively) recent times, from one where taxpayers (and advisers) created and controlled the data on their tax affairs and determined in the first instance which information they would insert into their tax return to a situation where increasing amounts of data are created (or at least selected) by the ATO from data provided by third parties rather than by the taxpayer.

This change undermines the original rationale for imposing the burden of proof on taxpayers – that they would have better access to the information about their tax affairs than the ATO. It is time to update the burden of proof provisions to recognise the new realities and transfer the burden of proof to the ATO, applying the contra proferentem principle.

The current position is both illogical and unfair, and if one applies the same rationale for the original allocation of the burden of proof to the modern situation,(ie that the onus should lie on the party contributing the most significant amount of data) the argument for transferring the burden of proof to the ATO is overwhelming – albeit we may need to tread water for a (short?) time while the ATO expands its pre-filling abilities to non-individuals to an appropriate level.

The system needs to recognise that the burden of proof provisions have not kept up with the implications of technological change, and that individual taxpayers still bear the burden of proof, even though it is now the ATO which inserts and validates more and more data into their tax returns.

It is past time to update the current burden of proof provisions.

^* Robin Woellner is an Adjunct Professor in the School of Law at James Cook University and the School of Tax and Business Law at UNSW. He would like to thank Emeritus Professor Stephen Graw for his assistance in finalising this paper for publication.

^[1] The list of categories of required information publications is too lengthy to set out here but see the list in the text accompanying n 17; see also ATO ‘Data matching’ https://www.ato.gov.au/about-ato/commitments-and-reporting/information-and-privacy/data-matching.

^[2] Thus, s 14ZZO provides that ‘In proceedings on an appeal under section 14ZZ to a court against an ATO objection decision, the appellant [i.e. the taxpayer] must prove [on the balance of probabilities] that: ...’

‘If the taxation decision concerned is an assessment – that the assessment is excessive or otherwise incorrect and what the assessment should have been; or

In any other case – that the taxation decision should not have been made or should have been made differently.’

^[3]FC of T v Ross [2021] FCA 766 [46].

^[4] See Trautwein v FC of T [1936] HCA 77; (1936) 56 CLR 63, 87.

^[5] FC of T v Ross (n 3) [46].

^[6] There is a useful summary of relevant principles in Ross (n 3) [48], points 1-9 (Derrington J).

^[7] However, the courts have held that it is not necessary for the taxpayer to determine their taxable income to the exact dollar – Nguyen v FC of T [2018] FCA 1420; (2018) 265 FCR 355 – but they must establish an amount which is ‘more nearly right’: Trautwein v FC of T (1936) 56 CLR 88, cited in Ross (n 3) [65] (Derrington J)]

^[8] Thus, for example, in Ross (n 3) at [56] Derrington J) stated that the ‘... possibility of Mr Wang demonstrating errors in the asset betterment methodology did not advance Mr Wang’s case as it fell short of what was required to discharge his onus under s 14ZZK(b)(i) ...’. The alternative ground – that the assessment was ‘otherwise incorrect’, is rarely discussed.

^[9] FC of T v Dalco (1990) 168 CLR 614; see also Bosanac v FC of T [2019] HCA 41; (2019) 93 ALJR 1327.

^[10] Though it has been said that the general legal principles relating to assessments under ss 166 and 167 ITAA 36 are the same: Ross (above) [22].

^[11] FC of T v Ross [201] FCA 766,

^[12] Ross, above, (n 3) at [48], citing Gashi [77]-[79]; [77]-[79]; Woellner and Zettle (sic) ‘satisfying the Taxpayer’s Burden of Proof in Challenging A Default Assessment – The Modern Labours of Sisyphus?’ [2014] JALaw TA 11.

^[13] From among very numerous cases, see, for example, Confidential v FC of T [2014] AATA 32 [89] (Forgie DP); Healey v FC of T (No 2) [2012] ATC 20-365 [56]-[102]; Healey v FC of T (No 2) [2012] ATC [14] Healey v Federal Commissioner of Taxation (2012) FCA 269.

^[14] That is, they continue to appropriately balance the ability of the ATO and taxpayers to access relevant information.

^[15] Categories of third parties required to provide specified information by set times and sources and content are listed in, for example, ATO ‘Data matching’, https://www.ato.gov.au/tax-and-super-professionals/for-tax-professionals/prepare-and-lodge/tax-time/before-you-lodge/data-matching (cf https://www.aph.gov.au/Parliamentary_Business/Committees/House/Former_Committees/Tax_and_Revenue/Taxpayerengagement /Report_1). ATO data matching usually occurs in the years after pre-filling.

^[16] See also ATO ‘Data matching’, ‘Expanding our data matching: https://www.ato.gov.au/tax-and-super-professionals/for-tax-professionals/prepare-and-lodge/tax-time/before-you-lodge/data-matching; While the categories of required information provision may seem extensive, there are various gaps in the information required to be provided to the ATO – including, for example, CGS, FSI and cash income.

^[17] In broad outline – the detailed categories are set out in, for example, the ATO ‘Pre-filling 2023’ Report: see https://www.ato.gov.au/tax-and-super-professionals/for-tax-professionals/prepare-and-lodge/in-detail/pre-filling-reports/pre-filling-2024.

^[18] The ATO draws data from, for example, the ‘MyDeductions’ app, which collates data on work-related expenses, interest and dividend deductions, gifts or donations, the cost of managing a taxpayer’s tax affairs, sole trader expenses, business income and other deductions. The ATO notes that where the total deduction for items D1 to D15 was less than $300, ‘the ... Pre-filling report will show details from the prior year’s tax return, for example from work-related clothing, laundry and dry-cleaning expenses, other work-related expenses, and gifts or donations.

^[19] Sweden allows taxpayers to accept a pre-filled return by the internet, phone, SMS or on paper – see:

https://www.aph.gov.au/Parliamentary_Business/Committees/House/Former_Committees/Tax_and_Revenue/Taxpayerengagement/Report_1.

^[20] OECD Forum on Tax Administration: Taxpayer Services Subgroup – Information Note: ‘Third Party Reporting Arrangements and pre-filled Tax Returns: The Danish and Swedish Approaches, 22 January 2008, [9]: https://www.aph.gov.au/Parliamentary_Business/Committees/House/Former_Committees/Tax_and_Revenue/Taxpayerengagement /Report_1 [2.27].

^[21] ‘Pre-filling 2023’ Treasury.gov.au; see also ibid.

^[22] https://www.ato.gov.au/tax-and-super-professionals/for-tax-professionals/prepare-and-lodge/tax-time/before-you-lodge/data-matching; see also ‘Improving tax compliance – enhanced third party reporting, pre-filling and data matching’, Australian Government, Discussion Paper, February 2014, and OECD, Information Note’: ‘Using Third Party Information Reports to Assist Taxpayers Meet Their Return Filing Obligations – Country Experiences With the Use of Pre-populated Personal Tax Returns, March 2006’: https://www.scribd.com/document/354124709/36280368 .

^[23] Ibid.

^[24] An Early analysis by Chris Evans and Binh Tran-Nam, ‘Pre-Filled Personal Income Tax Returns: Evidence from Australia’103^rd Annual Conference on Taxation171-173, National Tax Association Proceedings suggested that any reductions tax compliance costs of those personal taxpayers who do not use tax agents would not be significant, while the compliance costs of those who do would be very marginal.

^[25]https://www.aph.gov.au/Parliamentary_Business/Committees/House/Former_Committees/Tax_and_Revenue/Taxpayerengagement/Report_1/section?id=committees%2freportrep%2f024169%2f25967

^[26] Kevin Brookes in collaboration with Pascale Dery, (2018) Economic Notes, Taxation Series, August 2018 ‘Should the Government pre-fill Your tax return?’

^[27] Ibid

^[28] Ibid, 1-4, see Table 1.

^[29] OECD (n 20), Summary. However, the ATO would need to incur costs in any event for data-matching, post lodgement checking and audit and related activities.

^[30] Information may be incomplete because of issues in relation to timing, matching or validation: ATO ‘About pre-filling’, https://www.ato.gov.au/tax-and-super-professionals/for-tax-professionals/prepare-and-lodge/pre-filling-service/about-pre-filling.

^[31] Brookes and Dery (n 26).

^[32] Accordingly, Ddawati Ibrahim and Jeff Pope, speculate that under a pre-fill system ‘There is a possibility that, without [a tax return] being issued, taxpayers became more lackadaisical’; see also: ‘The Viability of a Pre-Filled Income Tax Return System for Malaysia’, The Journal of Contemporary Issues in Business and Government (2011) 17(2), 85-101.

^[33] As Evans and Bihn Tran-Nam (n 24) put it, ‘as more and more pre-filling data become available to e-tax users, and as they come to trust and use that information more and more, there may be a danger that the role of the personal taxpayer as a self-assessor becomes less well-defined’.

^[34] Brookes and Dery (n 26), 3.

^[35] By comparison, a 2015 OECD Review found that almost half of the revenue bodies then surveyed applied pre-filling, and that Estonia, Finland, Iceland, Lithuania, Norway, Peru, Portugal, South Africa and Sweden provided full or near-full coverage of pre-filled information for personal taxpayers and 11 jurisdictions applying a ‘deemed acceptance’ of the pre-fill return if the taxpayer has not responded within after a set period. See the list of categories and details required in: ‘Pre-filling 2009-11’, https://www.ato.gov.au/tax-and-super-professionals/for-tax -professionals/prepare-and-lodge/in-detail/pre-filling-reports/pre-filling-2009-2011.

^[36] In broad outline – the detailed categories are set out, for example, in the ATO ‘Pre-filling 2023’ Report’: https://www.ato.gov.au/tax-and-super-professionals/for-tax-professionals/prepare-and-lodge/in-detail/pre-filling-reports/pre-filling-2022-2023.

^[37] Josh Needs ‘Use data prefills or raise a red flag’ Accountants Daily 3 June 2022 citing ATO Assistant Commissioner Tim Loh. See https://www.accountantsdaily.com.au/tax-compliance/17122-use-data-prefills-or-raise-a-red-flag-ato-warns.

^[38] ATO Assistant Commissioner Tim Loh stated that ‘Around 80% of taxpayers with rental income claimed a deduction for interest on their loan, and this is where we’re seeing mistakes. For example, you can’t refinance an investment property to buy personal items, like a holiday to Europe or a Tesla, then continue to claim the interest expenses as a tax deduction’: https://www.ato.gov.au/media-centre/get-your-rental-right-this-tax-time

^[39] From 1 July 2023, the ATO has been receiving data on income under the Sharing Economy Reporting Regime: from electronic distribution platforms providing taxi services, ride-sourcing, and short-term accommodation; from 1 July 2024, all electronic distribution platforms are required to provide income data: see Loh (n 38).

^[40] See ‘ATO expands data matching to ensure fair play’: https://www.ato.gov.au/media-centre/ato-expands-data-matching-to-ensure-fair-play.

^[41] Ibid.

^[42] Jason Kerr suggests adoption of a ‘hybrid’ model containing elements of both pre-filled returns and a reduced filing system: Jason Kerr, ‘Tax Return simplification: risk key engagement, a return to risk?’ (2012)10(2) eJournal of Tax Research 465-82’. See also ATO ‘Data matching’, ‘Expanding our data matching; https://www.ato.gov.au/tax-and-super-professionals/for-tax-professionals/prepare-and-lodge/tax-time/before-you-lodge/data-matching.

^[43]https//www.aph.gov.au/Parliamentary_Business/Committees/House/Former_Committees/Tax_and_Revenue/Taxpayerengagement /Report_1.

^[44] See, for example, https://www.ato.gov.au/individuals-and-families/your-tax-return/how-to-lodge-your-tax-return/lodge-your-tax-return-online-with-mytax/pre-filling-your-online-tax-return. In 2022, the ATO introduced an ‘enhanced prefill’ policy, under which taxpayers cannot change prefilled bank interest unless the client can provide evidence as to why the amount should be changed – the ATO noted that this resulted in a 74% decrease in compliance activities after lodgement.

^[45] See Needs (n 37) noting that Loh also warned that taxpayers and advisers should use ATO pre-fills to avoid the or risk of raising a ‘red flag’, with failure to wait for a pre-fill or altering form data running the risk of ‘putting clients under scrutiny’..

^[46] Needs (n 37).

^[47] Ibid.

^[48] Kerr (n 42) suggests a ‘hybrid’ model containing elements pre-filled returns and a reduced filing system.

^[49] More detailed comparisons with overseas tax systems will be the subject of future papers.

^[50] OECD (n 20) [9].

^[51] Reporting bodies are listed in ibid [20]-]21]

^[52] In Denmark, reporting on deductions ‘... obligations typically extend to pension contributions, union fees, unemployment insurance, and interest on home mortgages’. Ibid, [26].

^[53] In Denmark, for example (cf Sweden, above), if a taxpayer does not respond within a prescribed period, this is deemed to constitute acceptance of the return. The acceptance rate was some 65% in Sweden, and 72% in Denmark.

^[54] OECD (n 20) [44]-[46].

^[55] See nn 37 and 38.

^[56] In 2022-2023, the ATO reported that some 98.7% of individual tax returns were lodged online: see Commissioner of Taxation Annual Report 2022-2023, Table 7.21, p.237.

^[57] See n 15.

^[58] In this context, it can also be argued that in a real sense, the ATO ‘creates’ data when it decides which data it receives from third parties it will regard as having a sufficiently ‘high degree of confidence’ to merit inclusion in a taxpayer’s pre-filled return. This decision involves an open-ended value judgment in a context where the ATO could potentially be subject to competing pressures (for example, pressure from a federal government seeking to maximise revenue returns). This might raise questions as to whether this is an appropriate role for the ATO.

^[59] Literally ‘against the offeror’.

^[60] It seems unlikely that many taxpayers would be aware of the burden of proof requirements.

^[61] The ‘tipping point’ would need to be set in advance by the parties and could be, for example, 51% or any other agreed percentage – or other discrimen.

^[62] Past experience suggests that the ATO will in the future be able to access data from non-individuals effectively. Some other countries seem to be heading in that direction already – see the text accompanying n 50.

^[63] My thanks to Michael Bersten, who brought these issues to my attention.