Australia's Future Tax System

Final Report: Detailed Analysis

Chapter G: Institutions, governance and administration

G5. Monitoring and reporting on the system

G5–2 Managing, measuring and reporting on tax expenditures

The main objective of any tax system is to raise revenue to fund government activities. However, the system also provides a government with the opportunity to achieve objectives directly, by reducing taxes to encourage certain activities and to direct assistance to particular groups. These forms of assistance are known as tax expenditures.12 They can be provided in many ways, including through exemptions from tax, tax deductions, tax offsets, concessional rates of tax, a change in the timing of a deduction or the deferral of a tax liability. Not all concessions delivered through the tax system are tax expenditures. For instance, some refundable tax offsets are classified as spending programs, even though they are administered through the tax system.

Tax expenditures provide many billions of dollars of relief from taxes, with significant effects on the budget and the economy. They reduce revenue that, if collected, would have been available to fund spending programs to meet similar objectives. Accordingly, their net effect on the budget is similar to that of spending programs. Assuming the case for government assistance has been made, a decision for policy-makers is whether to deliver that assistance by making a direct payment (such as a grant) or conferring a tax concession.

High standards of transparency and accountability should apply to all forms of government expenditure to engender greater community trust in, and therefore engagement with, the tax and transfer system.

Chart G5-1: Tax expenditures and direct expenditures in 2007–08

Chart G5-1: Tax expenditures and direct expenditures in 2007–08

Source: Tax Expenditures Statement 2008.

Principles

Tax expenditures should ideally be subject to the same levels of transparency and accountability as equivalent spending programs. Without such transparency and accountability, programs should not be delivered as tax expenditures, unless there is a clear countervailing benefit in terms of efficiency, equity, complexity, sustainability and policy consistency.

Comparing tax expenditures and equivalent spending programs

Despite their resemblance, tax expenditures differ from spending programs in a number of ways. For instance, tax expenditures often receive Parliament's attention only at the time they are introduced, and they are subject to a less comprehensive management and reporting framework than spending programs. These differences, examined in more detail below, are often the source of concerns about the transparency of tax expenditures.

Tax expenditures and direct spending programs are often designed in quite different ways. In particular, the design of tax expenditures is often constrained in practice by the policy, legislative and administrative framework of the tax system. These constraints can significantly affect the efficiency, equity, complexity, sustainability and policy consistency of tax expenditures.

Efficiency

Like direct expenditure programs, tax expenditures can impose efficiency costs by encouraging taxpayers to undertake tax-favoured activities at the expense of other activities.

Some tax expenditures aim to correct market failures and so may improve efficiency by encouraging activities that would otherwise be underprovided. However, governments aim to encourage such activities without providing a windfall gain to people who would undertake them without the tax expenditure. It can be difficult for governments to set the level of a concession that encourages the favoured activities to the intended extent. This can also be true of spending programs. However, the fact that tax expenditures are typically claimed at the end of a financial year means there is less opportunity for the government to monitor the effect of the tax expenditure during the year to ensure it is having the desired effect.

On the other hand, use of tax expenditures rather than spending programs can reduce the need for direct government supervision of a policy, thereby saving administration costs. For example, there may be administrative savings from using the tax system rather than establishing a new bureaucracy to implement a spending program. However, in some cases the costs of a tax expenditure may simply not be as apparent. While establishing a new agency will have obvious budgetary and administrative costs, the extra costs borne by the tax administrator may not be as obvious.

In some cases, the tax system is a more administratively convenient way of delivering assistance than an equivalent spending program. However, the tax system may not be as efficient for the types of assistance that depend on characteristics not commonly reported in tax returns.

Finally, there is a question about the ability of the tax authorities to deliver non-tax programs efficiently. Naturally, tax authorities have built their expertise around the collection of taxes. The proliferation of tax expenditures has increased pressure on tax authorities to develop the ability to deliver other programs. It is important to consider whether the tax authority is properly equipped to perform this role, and also whether this impedes the ability of the tax authority to perform its core revenue collection function.

Equity

Many tax expenditures have been introduced to improve the fairness of the system. However, some tax expenditures actually reduce the equity of the system by directing benefits disproportionately to people on higher incomes.

For example, some tax expenditures directed towards individual taxpayers are delivered through income tax deductions. Progressive rates of the personal income mean these tax deductions have a greater value for those on higher incomes.

This is not an inherent weakness of tax expenditures as a policy tool. Rather it is a consequence of delivering particular tax expenditures as personal tax deductions, when they could have been provided as tax offsets (which have the same value to all recipients).

Complexity

Tax expenditures can add complexity to the tax system. They complicate the law and create additional choices for people. They can generate unintended opportunities for tax planning.

While providing a benefit through the tax system may reduce program administration costs, in practice it may impose extra compliance costs on taxpayers who need to understand how the concession applies. It may also impose costs on taxpayers who do not benefit from the tax expenditure, particularly in a system of compulsory filing. Moreover, as new tax expenditures are added, complexity increases because taxpayers need to understand the compounding number of interactions between the many rules.

Most tax expenditures could be delivered as direct expenditures. This would improve the tax system but would not necessarily reduce the overall level of complexity imposed on society. In the end, the focus should be on total complexity, regardless of whether that complexity arises within or outside the tax system.

Sustainability

Tax expenditures should be considered in light of revenue sustainability. Policies need to be affordable over the longer term, particularly in light of the demographic challenges facing Australia.

Tax expenditures can be difficult to contain. They can lead to erosion of the tax base as different groups put their case for further concessions. Once the system features a number of tax expenditures, the case for further tax expenditures is strengthened. In other words, the special treatment of one sort of taxpayer or activity tends to weaken the benchmark treatment against calls for similar concessions for other taxpayers or activities.

Tax expenditures can make it difficult to appreciate the extent of government intervention. For instance, they can reduce both tax collections and direct expenditures, making government appear smaller than it would if the same policies were pursued through direct expenditures. Once a tax expenditure is established, it can be politically difficult to remove it because that can be portrayed as a tax increase.

The proliferation of tax concessions may also threaten the integrity of the system. Voluntary compliance of taxpayers is a valuable aspect of our tax system. In particular, the collection of income tax under a system of self-assessment depends heavily on the community's trust in, and commitment to, the system. Tax expenditures can undermine this trust, especially when they are seen to be unfair or complex. As noted above, the design of many tax expenditures, and their lack of transparency and accountability, can contribute to these perceptions.

Around 320 tax expenditures were identified in the Tax Expenditures Statement 2008. In the past five years, around 50 new tax expenditures have been added to the federal tax system. A similar number have been identified as tax expenditures for the first time during this period. The value of tax expenditures has risen from an estimated $50.2 billion (5.3 per cent of GDP) in 2004–05 to an estimated $73.7 billion (7.1 per cent of GDP) in 2007–08.13 While tax expenditures are not projected to continue to grow at this pace, the figures highlight the significance of tax expenditures and the restrictions they impose on the government's ability to fund services.

Chart G5-2: Australian government tax expenditures

Chart G5-2: Australian government tax expenditures

Source: Tax Expenditures Statement 2008.

Policy consistency

A final design principle is that the tax and transfer system should be consistent with broader policy objectives. This suggests that all policies should be considered in an integrated way, and that it is their combined impact that matters. This is particularly relevant when assessing the role of tax expenditures, since the justification for many of them lies in other economic and social policy objectives.

Findings

While tax expenditures and direct spending programs are conceptually similar they are often designed in quite different ways. The design constraints on tax expenditures can significantly affect their efficiency, equity, complexity, sustainability and policy consistency.

Determining whether the benefits of tax expenditures justify their costs depends on effective monitoring and scrutiny. Tax expenditures are currently subject to less comprehensive management and reporting than spending programs. This hampers the effective supervision of individual tax expenditures and means that, in many cases, it is not possible to work out whether objectives are being achieved.

Making tax expenditures more transparent and accountable

Recommendation 135:

The Australian government should ensure that the rules governing the development of the Budget encourage trade-offs between tax expenditures and spending programs. Budget decision-making processes should measure and treat tax expenditures and spending programs symmetrically, to ensure that there is no artificial incentive to deliver programs through one mechanism rather than another.

Recommendation 136:

The government should introduce legislation to amend the Charter of Budget Honesty Act 1998 to recognise the publication of detailed information about tax expenditures in a Tax Expenditures Statement separate from the Mid-Year Economic and Fiscal Outlook (MYEFO). However, the Tax Expenditures Statement should continue to be released by the end of January in each year, or within six months of the last Budget, whichever is later.

Recommendation 137:

The government should ensure that reporting standards are independently developed for the identification and measurement of tax expenditures in the Tax Expenditures Statement. In addition, the standards should establish a basis for reporting the broader economic and distributional effects of tax expenditures in the periodic Tax and Transfer Analysis Statement (see Recommendation 132).

Recommendation 138:

The Council of Australian Governments should examine the ways in which the States could uniformly report tax expenditures annually according to the independent standards developed under Recommendation 137.

Encouraging trade-offs between tax expenditures and spending programs

The Budget is the government's key decision-making process for revenue policy. It is therefore the most important process by which new tax expenditures are created and existing tax expenditures can be reviewed.

Under the Budget process, tax expenditures are settled after spending measures. The two are not usually examined together. This means that tax expenditures are not directly compared with other policy priorities and new spending proposals. There are no formal processes to ensure that tax expenditures are prioritised against other spending or to assess the efficiency of a tax expenditure in achieving outcomes. This increases the risk that tax expenditures are not properly coordinated with spending programs in the same policy area.

Ministers may, in conjunction with the Treasurer, propose new tax concessions on the basis that the reduction in revenue is offset by savings from within the Minister's portfolio. In this way tax expenditures can be controlled at the policy development stage by ensuring that the cost of any new concession is counted against the relevant portfolio budget and that offsetting savings are required in the same way as for spending programs.

However, ministers are not usually able to claim as savings any increases in revenue that might flow from the removal of an existing tax expenditure. In the past, this has tended to discourage the replacement of tax concessions with equivalent spending programs.

A more symmetrical treatment of tax expenditures and spending programs as part of the Budget process would encourage trade-offs between them and would help to ensure that policy objectives are pursued at least cost (see Recommendation 135). To do this, however, requires that expenditures and tax expenditures are measured on a consistent basis.

The establishment of 'spending rules' that limit the size or growth of government spending can have the perverse incentive of encouraging new tax expenditures, unless there is a similar constraint on revenue policy. Such rules create incentives to use tax expenditures as substitutes for government spending. This was a problem Sweden experienced when it imposed a spending cap with no restriction on tax expenditures (OECD 2009a). Australia should be mindful of this example and ensure that the restraint applied to spending proposals is also applied to tax expenditures.

Reporting tax expenditures more effectively

In contrast to direct government spending, which is generally scrutinised during the annual Budget process, tax expenditures often receive attention only at the time they are introduced. Systematic reporting of tax expenditures is therefore necessary to ensure they receive a similar degree of scrutiny as direct expenditures. This also makes it easier to compare tax expenditures and direct expenditures.

Australian governments have published annual estimates of tax expenditures since 1980, initially as an appendix to the Budget. The first separate Tax Expenditures Statement, providing detailed estimates of tax expenditures and the associated benchmarks, was published in 1986. The publication of this information became a legislative requirement under the Charter of Budget Honesty Act 1998.

An element of the framework established by the Act was that the Mid-Year Economic and Fiscal Outlook (MYEFO) would include detailed estimates of both tax expenditures and spending programs, thereby enhancing the scrutiny of both forms of expenditure. However, MYEFO estimates of tax expenditure have not yet been presented in a disaggregated form. Instead, this information is published separately in the Tax Expenditures Statement. MYEFO is released by the end of January in each year, or within six months of the last Budget, whichever is later. The separate Tax Expenditures Statement has also been released within this timeframe.

The purpose of MYEFO is to update key information in the most recent Budget. It provides an update on the government's fiscal and revenue strategy, rather than a comprehensive account of all measures. So, even if detailed estimates of tax expenditures could be produced for MYEFO, it would be difficult to compare them against spending programs in any detailed way.

Including fully detailed tax expenditure estimates in MYEFO would significantly change its focus as an update to the government's fiscal and revenue strategy and could delay its release. A better means for managing tax expenditures is by ensuring they are examined in the same way as spending programs in the Budget process. Detailed estimates of tax expenditures need to be prepared far enough ahead of the Budget to allow them to inform government decisions. The Tax Expenditures Statement should continue to be released within the same timeframes as MYEFO, though not necessarily at the same time. The Charter of Budget Honesty Act 1998 should be amended accordingly (see Recommendation 136).

Identifying tax expenditures

In order to identify a tax expenditure, the tax treatment that would normally apply (the benchmark) needs to be identified.

Not all concessional elements of the tax system are classified as tax expenditures. This is because some concessions are considered to be structural elements of the tax system and are incorporated in the benchmark. For example, the personal income tax system includes a progressive marginal tax rate scale, which results in individuals on lower incomes paying a lower marginal rate of income tax than those on higher incomes. This arrangement is a structural design feature of the Australian tax system and is therefore not identified as a tax expenditure. There may be different views on which structural elements to include in the benchmark. These benchmarks can vary over time and can sometimes be perceived as arbitrary.

The original concept of a tax expenditure includes only some of the concessional features of the tax system provided under the law. However, many tax benefits arise as a consequence of administrative practice or through non-compliance with the law. Some consider that these 'benefits' should be reported in the same way as other tax expenditures (ANAO 2008).

The purpose of reporting tax expenditures is so the community can understand how the tax system affects the economy and society more broadly. Benchmarks should allow an objective evaluation of the effects of government policy, rather than represent that policy. For example, if a tax concession is set up to assist a particular industry the benchmark should not incorporate this objective, but should provide a basis for identifying and valuing the concession. This allows the community to judge whether this form of assistance is appropriate.

Currently, many of the most important economic and distributional effects of taxes are incorporated in the benchmarks and so are not reported in the Tax Expenditures Statement. As noted above, a separate and broader Tax and Transfer Analysis Statement could include this kind of information about structural tax features. Even if this information is not reported annually in the Tax Expenditures Statement, the benchmark should be defined according to transparent and independently established standards.

The Charter of Budget Honesty Act 1998 requires budget reporting to be undertaken against a set of external standards. External standards are also supposed to apply to the reporting of tax expenditures, but no such standards exist. The development of these standards would improve the integrity of the process. They would provide greater consistency in the identification and measurement of tax expenditures, which would allow more reliable examination of trends over time.

Consideration needs to be given to whether an existing body should develop these standards or whether a new body should be established. One option the government might wish to consider is the establishment of an academic advisory panel that could independently develop standards for identifying and measuring tax expenditures. Whichever option is adopted, the body charged with the task should be equipped with the necessary skills and experience and should operate independently (see Recommendation 137).

Measuring tax expenditures

Unlike direct spending by the government, tax expenditures represent the notional cost to government of not collecting revenue that would otherwise be collected. These notional costs can be difficult to estimate, and the estimates can sometimes be misinterpreted as the amount of revenue that could be raised if the tax expenditures were abolished. For these reasons, tax expenditure estimates need to be treated with some caution.

The Tax Expenditures Statement estimates the value of tax expenditures using the 'revenue forgone' approach. This method is seen as the most reliable approach for estimating the level of assistance the tax system provides to taxpayers. Most other OECD countries also use this approach.

The revenue forgone approach calculates the benefit of a tax expenditure to taxpayers, rather than the budgetary cost of the expenditure. Estimates calculated by the revenue forgone approach identify the financial benefit to taxpayers of receiving a tax expenditure relative to taxpayers that do not. It does not necessarily follow that there would be an equivalent increase to government revenue from abolishing the tax expenditure. This is largely because of changes in taxpayer behaviour that removing the tax expenditure would cause (for example, removing one concession may result in increased use of others).

The 'revenue gain' approach has sometimes been proposed as an alternative to the revenue forgone approach in order to produce tax expenditure estimates that are more comparable to budget revenue estimates (ANAO 2008). This would directly measure how much revenue would increase if a concession were removed. It involves making assumptions about the way taxpayers would respond to policy changes. It also requires assumptions about the order in which tax expenditures are removed. This means there are considerable difficulties in preparing such estimates for all tax expenditures in the Tax Expenditures Statement. However, the Tax Expenditures Statement 2008 included revenue gain estimates for some major tax expenditures.

The revenue gain approach does not necessarily reflect the value of the concession to taxpayers. For instance, where an activity is highly sensitive to a concession, the increase in revenue from removing the tax expenditure could be very small. In these cases, revenue gain estimates give the impression that the tax expenditure has little impact, when in reality the recipients derive significant benefits. However, the revenue gain approach is useful when reviewing a tax expenditure since it indicates the revenue that could be realised for government if the expenditure were abolished. Revenue gain estimates for significant tax expenditures should continue to be published in the Tax Expenditures Statement.

A third approach to measuring tax expenditures is the 'outlay equivalence' approach. This approach estimates how much direct expenditure would be needed to provide a benefit to a recipient — assuming the payment is subject to the usual tax treatment for that type of income —that is equivalent to the tax expenditure.

The outlay equivalence method has the advantage of estimating tax expenditures on the same basis as spending programs, which may allow a better assessment of their comparative merits. Outlay equivalence estimates are likely to be most useful when policy-makers are considering whether to deliver a program as a tax expenditure or a spending program.

These estimates can differ significantly from estimates produced under a revenue forgone approach, which focuses on the annual effect of a concession on revenue collection. The Tax Expenditures Statement shows the annual effects of tax expenditures and so makes no distinction between concessions that defer tax and those that reduce tax. In effect, a tax deferral is a loan made by the government to a taxpayer and the value of the loan is the interest concession. For instance, the effect of accelerated depreciation is to allow taxpayers to claim larger deductions in the early life of an asset, and lower deductions in the latter part of the asset's life. Accelerated depreciation does not increase the size of the deductions that can be claimed overall, but it does bring them forward. This bring-forward represents an interest-free loan made by the government to taxpayers. Using the revenue forgone approach, the Tax Expenditures Statement shows the lower taxes paid in early years and the higher taxes paid later, rather the interest value of the deferred taxation. In contrast, the outlay equivalence method would estimate the value of the interest-free loan. This type of information about tax expenditures should be presented in the periodic Tax and Transfer Analysis Statement (see Recommendation 132).

Reporting State tax expenditures

In order to give a comprehensive sense of the level of government assistance provided through the entire tax system, tax expenditures need to be measured for all taxes. The Australian government currently reports tax expenditures across its main taxes. However, there is no comprehensive or consistent reporting of tax expenditures by the States. In particular, the benchmarks used by States differ significantly, so it is not possible to make a direct comparison of tax expenditures between jurisdictions.

To remedy this, reporting standards should be developed and applied across the range of State taxes in a uniform and thorough way. The Council of Australian Governments should examine the ways in which the States could uniformly report tax expenditures annually (see Recommendation 138).


12 The focus of this section is on those tax expenditures that reduce tax liabilities. However, the tax expenditure concept is defined as being any deviation from the 'normal' tax benchmark, and so also includes aspects of the tax system that increase tax liabilities. These 'negative tax expenditures' might be thought as being equivalent to a fee or an additional tax.

13 Tax Expenditures Statements, various years.