Previous Next

Agenda, Volume 16, Number 4, 2009

A ‘no-returns tax system’ for Australia: Some inconvenient facts

Sinclair Davidson[1]


It has been proposed that Australia adopt a ‘no-returns’ income tax system. One consequence of such a reform would be the standardisation, or even elimination, of work-related expense deductions. Using ATO data it is shown that in introducing a no-returns tax system and eliminating work-related expense deductions, the average Australian (making a claim) would save $268 on accountant fees but forgo $1860. In addition eliminating itemised work-related expense deductions would have a regressive impact.


One of the issues canvassed in the Australia’s Future Tax System Consultation paper (the ‘Henry Review’) is the simplification of tax administration (2008: 169–181). Individual Australian taxpayers are presently required to lodge an annual tax return. In addition, many taxpayers use accountants or tax agents to lodge their returns. This situation suggests two possible reforms. One reform would be to relax the requirement that all individual taxpayers lodge a return (generally described as a ‘no-returns tax system’; see Gale and Holtzblatt (1997)). A second reform would be to simplify the personal tax system so that individuals would not normally require professional tax advice in order to comply with their tax obligations (‘tax simplification’).

At a conceptual level these two types of reform are distinct; at a practical level, however, they are closely related. For relaxing the requirement that all individual taxpayers lodge a tax return would also require significant standardisation of deductions. In his 2004 discussion of annual tax returns, Chris Evans (2004: 180) stressed this important point: ‘Without annual filing, the taxpayer would lose the ability to claim expenses and so would lose the all-important refund.’[2]

This linking of the two types of reform is also suggested by the Henry Review (2008: 176–7) itself (emphasis added):

The ATO currently pre-fills some data into individual’s electronic income tax returns, making it easier for many individuals to complete their returns. With appropriate policy changes it may be possible to increase the amount of pre-filled data.

For example, introducing a standard tax deduction in place of work-related expenses and replacing the tax deductions for eligible gifts with a co-contribution to gift recipients would potentially mean that some individuals would need only to confirm the data in their pre-filled tax return at the end of an income year. Greater at-source withholding, complemented by policy settings that obviate the need for further assessment of tax, might further reduce the need for individuals to lodge returns.

As recently as 21 October, Steve Lewis (2009), writing in the Sydney paper The Daily Telegraph, suggested that the Henry Review would make such a recommendation:

Dr Henry’s reforms would result in taxpayers receiving a one-page summary from the Australian Tax Office which would include a standard ‘deduction’ for necessary work expenses as well as salary details. If taxpayers were happy with the ATO’s calculation, they would tick the form and lodge it via the internet — and wait for their refund. Benefits include taxpayers not having to make the annual trek to tax agents.

The idea of abolishing some deductions, or at least standardising those deductions, and simplifying tax returns isn’t new. Chris Evans of the University of New South Wales made a similar argument in 2004, while Andrew Leigh of the Australian National University wrote on this in 2007. Many other countries already operate a (largely) no-returns individual tax regime. New Zealand and the United Kingdom, for example, are only two of 36 countries that have a variant of this system in place.[3]

But it is not immediately clear that Australia should forgo work-related expenses deductions in return for a no-return tax system. While the benefits are easily understood they may not be as large as is generally presumed. Furthermore, a change to a no-returns policy is not in itself a substitute for tax simplification, and should only be considered after simplification has already occurred. To the extent that a no-returns policy could generate additional fiscal illusion, it is possible that a no-returns policy would inhibit tax simplification.

In the next section, I set out the benefits and costs of a no-returns policy from a taxpayer perspective and in the third section I consider some Australian evidence that may inform our views of the benefits and costs of a no-returns policy.

I conclude that forgoing work-related deductions for a simplified tax-return system would not be in the financial interests of the taxpayers, and ‘buying them out’ would be very expensive for the authorities.

Benefits and Costs of a No-Returns Tax System

The primary benefit of a return-free system is the reduced tax compliance burden for those taxpayers who are able to use the no-returns tax system. In addition to the need for careful record-keeping, taxpayers expend time and effort when filling out tax returns.[4] Andrew Leigh (2007: 85) suggests that the time could, rather, be spent working, or even relaxing. His argument is that the time saved not filling out a tax return would be equivalent to an additional public holiday. This is based on the widely accepted estimate that the average person takes 8.5 hours to fill out their tax return (Tran-Nam et al. 2000). Of course, most Australians do not fill out their own tax return but employ an accountant or tax agent to do so on their behalf. The direct costs of tax compliance would include the record-keeping costs, opportunity cost of filling out the tax return, and the cost of employing an accountant to do so.

There are also indirect costs of filling out a tax return. There may be significant emotional costs associated with dealing with tax matters. Many taxpayers may feel anxiety when confronted by somewhat complex tax documentation and the associated explanatory notes. Even those taxpayers who wish to fully comply with the tax system’s demands may be anxious that they have committed some error or other. It is not clear, however, whether these costs are significant or how widespread they may be. To some extent they may be subsumed into the direct costs associated with employing an accountant.

An additional benefit revolves around tax simplification that may occur when introducing a no-returns tax policy. I return to this point below.

The costs of a no-returns tax system revolve around a shift in compliance costs away from individual taxpayers to their employers and the tax authorities. In particular the burden of accuracy is shifted onto the tax authorities. It is one thing to recognise that the ATO already has a lot of information about individual taxpayers; it is another to collate that information in a timely and credible manner for use in an annual process. At present the ATO has several years in which it can review a tax return and compare that to information it already holds. Under a no-return tax system, it would have a few months in which to undertake that task for the entire taxpaying population. While the legal obligation to provide accurate information to the ATO would remain unchanged — that is, the burden is on the taxpayer — the fact that the ATO could be sending taxpayers incorrect information could undermine tax morale throughout the economy. The information technology start-up costs for such a system are likely to be very high.

When comparing the setting up of such a system with countries that already have that system it is important to remember that many countries have had these systems in place for a long time. The administration of the tax system would constrain some aspects of complexity, and would have evolved as the complexity of the tax system evolved. By contrast, the introduction of such a system after the complexity of the tax system had evolved would involve substantial ‘reverse-engineering’ that adds to the implementation costs.

The introduction of a no-returns tax system could contribute to greater levels of fiscal illusion (see Davidson 2007). This occurs when misconceptions about the tax burden minimise taxpayer resistance to higher levels of taxation. For example, an argument against lowering taxes is that government spending may fall as a consequence. Yet the Australian experience is that tax revenue has increased even though taxes, especially on high-income earners, have declined over the past few years. According to ATO Statistics, in 1996–97 the top 25 per cent of taxpayers paid 60.8 per cent of net personal income tax while in 2006–07 the top 25 per cent of taxpayers paid 66 per cent of net personal income tax.[5] The suggestion here is that a no-returns tax system would create an ‘invisible’ tax system that was ‘out of sight and out of mind’.

There is a common argument that taxpayers like receiving refunds. Many economists are suspicious of this argument because it implies that taxpayers are irrational. As Evans (2004: 179) asks: ‘What rational person overpays in order to get something back at a later stage?’ There is a potentially rational explanation, however, for this behaviour. Refunds reduce the size of the tax base. It is generally accepted by most economists that the more comprehensive the tax base, the lower the tax rate needs to be to raise a given amount of revenue. Consequently, from an economic perspective most economists would argue for broad bases and low tax rates. Geoffrey Brennan and James Buchanan (1980), however, have argued that taxpayers do not always trust government to deliver lower tax rates and consequently have a preference for narrower bases (see, especially, chapter three). This type of argument can reconcile a taxpayer preference for refunds and rationality. In this view, taxpayers are happy to trade off a loss of interest in the present against being over-taxed in the future. It would require careful survey analysis or even experimental techniques to provide evidence for this argument, and I provide no such evidence.

Some Facts of Costs of Managing Tax Affairs and Work Expense Deductions

In this section I evaluate the arguments for and against a no-return tax system using Australian data and experiences. This is done from the perspective of taxpayers and not from an ATO or tax-collection perspective. The data are drawn from the ATO Tax Statistics available on the web. The arguments put forward by Australian proponents of this type of tax system (for example, Andrew Leigh) usually emphasise the convenience factor of not having to lodge a return. The counter-argument to this view is that no-return tax systems only operate for taxpayers who have relatively simple tax affairs.

In the 2006–07 financial year only 27.54 per cent of Australia’s 11.8 million individual taxpayers prepared their own tax return. The vast majority of Australians employed an accountant or tax agent to prepare their return. The average cost for preparing the tax return in that year was $268.21. Figure 1 shows the average cost of preparing a tax return broken down by income groups. It also shows the average work-related expenses tax refund by income group. The overall average work-related refund was $1860.96 in that financial year. In other words, if making a trade-off between a no-return tax policy and work-related deductions, the average Australian (making a claim) would be asked to save $270 and give up over $1800.

Figure 1: Average costs and average work claims


Source: ATO Tax Statistics (2006–07), author’s calculations

Table 1 shows the average ‘cost of managing tax affairs’ — an allowable deduction and reported in aggregate by the ATO — over the period 1996–97 to 2006–07. The second column shows the percentage of taxpayers making this particular claim, while the third column reports the percentage of taxpayers the ATO reports as using a tax agent. The large number of taxpayers employing a tax agent or accountant implies that the Australian tax system is quite complex.

Indeed in a February 2006 speech, the Tax Commissioner, Michael D’Ascenzo, admitted that understanding tax law was beyond the comprehension of normal people (quoted in Roskam 2006). In short, this implies that the tax system itself is complex and would require substantial simplification before reform to the tax administration could be considered. It is an open question as to how many of these individuals would switch to a no-return tax system.

Table 1: Cost of managing tax affairs

Average Fee

Proportion of taxpayers making claim for the cost of managing tax affairs

Proportion of Taxpayers using a tax agent













































Source: ATO Tax Statistics (2006-07), author’s calculations

What of the remaining taxpayers who currently prepare their own tax returns? What are the costs associated with their compliance? To calculate this cost I calculated the annual after-tax income for taxpayers in their income groups, and then translated that into an hourly rate (based on 24 hours in the day and 365 days in the year). I then multiplied that figure by 8.5 hours — the generally accepted number of hours taxpayers use to manage their tax affairs.[6] The ATO has reported that the average time taken to complete a tax return is only 5.9 hours. To the extent that the ATO figures are correct, rather than the generally accepted figure, the calculations reported below are likely to overstate the costs of completing a tax return. Across the 27.54 per cent of taxpayers who lodge their own returns, this all adds up to $88.3 million in 2006–07. (The overall average, not taking into account the income distribution, is $112.2 million in 2006–07). The average cost for these taxpayers was $38.50 in that year or $41.59 if I ignore the income distribution. Even if my calculations are out by a factor of three, it would still add up to only $265 million. By comparison, taxpayers claimed $1.4 billion in the costs of managing their tax affairs. The cost of preparing a tax return for the proportion of taxpayers that prepare their own returns is between 6 and 19 per cent of the cost of the rest of the taxpaying population. That, of course, doesn’t indicate that no effort should be made to further reduce this cost, but it does suggest that the benefits of doing so are likely to be low.

I now look specifically at the idea of trading off work-related expense deductions for a no-return tax system. Table 2 shows that the average claim has increased slightly over time, as has the percentage of taxpayers making work-related claims. At the same time, however, it is important to note that the economy grew very dramatically over the period 1996–97 to 2006–07. Increases in work-related expenditure are not unreasonable.

Table 2: Work-related expenditure

Total work-related expenses as % of total taxable income

Percentage of taxpayers making work-related claims


































Source: ATO Tax Statistics (2006–07), author’s calculations

In 2006–07, work-related expense deductions comprised 41.54 per cent of all deductions in dollar terms, down from a high of 58.70 per cent in 1999–2000. Similarly, they comprise about 80 per cent of all deductions that are claimed. In other words, work-related expense deductions comprise a very large component of the ATO workload. Yet it is quite clear when looking at the income distribution of claimants that many lower income individuals benefit from having these deductions.

Figure 2 shows that individuals earning between $23,000 and $70,000 make the most use of work-related deductions, with this type of deduction making up more than 50 per cent of all deductions in this income group. As a rule of thumb this income group can be described as earning between half and one and a half times the average income for a full-time employee. Higher-income taxpayers tend to have deductions for gifts and donations, interest and dividends and for non-employer sponsored superannuation. Figure 3 examines who actually makes work-related deduction claims by income group. The single largest group of taxpayers making work-related claims earn between $30,000 and $35,000, while the largest monetary value of claim is for individuals earning between $60,000 and $70,000. Looking at information such as this indicates that work-related claims substantially lower the effective tax rates for lower-income and average-income Australian workers. The graph also shows evidence that taxpayers use work-related deductions to reduce their nominal tax rates — the spikes at $70,000 and $150,000 closely match tax thresholds (in that year) from the 30 per cent to 40 per cent rate ($75,000) and the 40 per cent to 45 per cent rate ($150,000).[7] Any policy to remove work-related deductions is likely to be regressive.

Figure 2: Work-related deductions as percentage of total deductions


Source: ATO Tax Statistics (2006–07), author’s calculations

Figure 3: Who makes work-related claims?


Source: ATO Tax Statistics (2006–07), author’s calculations

Figure 4 confirms this view by calculating the cumulative proportion of taxpayers making a claim and cumulative dollar value of the work-related claims. As can be seen, over 50 per cent of all work-related claims are made by individuals earning less than $40,000, while 50 per cent of the dollar value of claims is made by individuals earning less than $50,000.

Figure 4: Cumulative work-related deductions


Source: ATO Tax Statistics (2006–07), author’s calculations

An argument that is often made is that itemised deductions could be replaced with a standardised deduction. It is quite likely that this would be very expensive; the deduction would have to reasonably compensate current claimants but also be accessible to those individuals who do not currently make work-related deductions. At the same time, a standardised deduction could be eroded by inflation (like the tax-free threshold has been eroded) or even eliminated altogether by a future government. A referee has suggested that a standardised deduction that is a function of employment income could be adopted in place of an itemised deduction. This proposal is superior to a standard one-size-fit-all approach, but nonetheless it does not take account of the regressive nature of the current work-related expense deduction.


The argument this paper alludes to is that simplification of the tax system itself must precede simplification of tax administration. Certainly the idea of trading off work-related deductions for a simplified tax-return system would not be in the financial interests of the taxpayers and buying them out would be very expensive for the authorities. Andrew Leigh (2007: 84) has argued in favour of a no-return tax system as follows:

Plenty of taxpayers may choose to forfeit deductions to which they are entitled in exchange for avoiding the hassle of filing a tax return. Indeed, the rise in tax revenue from these forfeited deductions is likely to outweigh any increase in administrative costs for the ATO in moving towards a system of pre-population.

The analysis presented in this paper, however, suggests that his argument is wishful thinking at best. Few taxpayers would want to give up their work-related deductions to save the cost of an accountant or tax agent.


Australia’s Future Tax System consultation paper (‘Henry Review’), December 2008, Commonwealth of Australia, available at:

Australian Tax Office 2009, Taxation Statistics 2006–07, available at:

Brennan, Geoffrey and Buchanan, James M. 1980, The Power to Tax: Analytical Foundations of a Fiscal Constitution, Cambridge University Press.

Davidson, Sinclair 2007, ‘Fiscal illusion: How big government makes tax look small’, Perspectives on Tax Reform (15), CIS Policy Monograph 81. Available at:

Evans, Chris 2004, ‘Diminishing Returns, The case for reduced annual filing for personal income taxpayers in Australia’, Australian Tax Review 33: 168–81.

Gale, William and Holtzblatt, Janet 1997, ‘On the possibility of a no-return tax system’, National Tax Journal 50: 475–85.

Leigh, Andrew 2007, ‘Taxpaying made easy’, Agenda 14: 81–6.

Lewis, Steve 2009, ‘Tax returns could be scrapped under radical new proposal’, The Daily Telegraph, 21 October; available at:, accessed 21 October 2009.

Atkinson, Anthony and Leigh, Andrew 2007, ‘The Distribution of Top Incomes in Australia’, Economic Record 83(262): 247–61.

Roskam, John 2006, ‘Tangled web of federal law’, Australian Financial Review, 10 February 2006: 82.

Tran-Nam, Binh, Evans, Chris, Walpole, Michael and Ritchie, Katherine 2000, ‘Tax compliance costs: Research methodology and empirical evidence from Australia’, National Tax Journal 53: 229–52.

[1] School of Economics, RMIT University, and the Institute of Public Affairs; I thank William Coleman, Andrew Leigh and two anonymous referees for comments on a previous version of this paper.

[2] Evans then raises the question, ‘whether there would be significant taxpayer resistance if … taxpayers lost the right to claim work-related expenses (and hence their annual refund)’.

[3] There are different types of no-return tax systems that operate around the world. The two major systems are known as the Tax Agency Reconciliation System and the Exact Withholding System. A hybrid of these two systems operates in the Nordic countries. See Gale and Holtzblatt (1997) for a general discussion.

[4] It is important to note that the need for record-keeping is invariant to whether a no-returns tax system is adopted or not.

[5] In correspondence Andrew Leigh argues that the tax share of the top 25 per cent of taxpayers has risen because their share of income has risen (see Atkinson and Leigh 2007 for further discussion). There may well be merit to this argument, but unravelling the various effects of tax policy will take this paper far beyond its more limited and modest objective.

[6] In the analysis I have made two important assumptions. I assume that individuals across various income categories value their time equally as a function of their income. This is a common assumption, but I am not convinced that it is an accurate reflection of actual behaviour.

[7] Unfortunately the ATO data do not coincide with the $75,000 cut-off for the 30 per cent tax rate.

Previous Next