Architecture of Australia's tax and transfer system
There is no direct measure of complexity in the tax‑transfer system
While there is a community view that the tax‑transfer system is complex, and the level of complexity has been increasing, it is very difficult to measure this complexity in an objective way. Proxies for complexity are often cited as evidence of the complexity of the tax‑transfer system. These include: the number of pages of legislation; the frequency with which it is amended; the proportion of taxpayers using tax agents to complete their tax returns; the time taken to prepare tax returns; the average level of customer contact required in the administration of transfers; the number of administrative rulings that are required to clarify the law; the number of reviews of administrative decisions and the number of disputes that are litigated.
These measures are at best only indicative of possible trends in complexity and each must be viewed with some caution. For example, while there is likely to be a positive correlation between the overall size of the income tax legislation and its complexity, this relationship does not always hold. In the late 1990s the Tax Law Improvement Project was established to rewrite the income tax law to make it easier to understand. The new drafting style, while clearer, better structured and shorter, did not involve any significant policy simplification. Similarly, the removal of more than 4,100 pages of 'inoperative provisions' from Australia's income tax legislation in 2006 improved its readability. However, the reduction in complexity was not commensurate with the reduction in the size of the law, as it did not reduce the number of operative rules or their complexity. The proportion of taxpayers using tax agents is another frequently cited indicator of complexity in the tax system. However, the use of tax agents may reflect taxpayers' assessments about the value of their time or an expectation of a better tax outcome if they use an agent.
Although there are weaknesses in each of these partial measures of complexity, collectively the various indicators in Chart 11.1 suggest the level of complexity in the tax‑transfer system may not have changed markedly in recent years. However, this does not provide any insights into whether the current levels of complexity and compliance costs are appropriate.
Chart 11.1: Partial indicators of complexity in the tax‑transfer system
- AAT means Administrative Appeals Tribunal.
Source: ATO, FaHCSIA and Australian Treasury estimates.
Only limited data are available on administration and compliance costs
An alternative approach to evaluating the impact of complexity in the tax‑transfer system is through the direct assessment of administration and compliance costs. The relationship between them is again not necessarily a strict one. However, administration and compliance costs are of interest in their own right, since they represent an important part of the efficiency cost of the tax‑transfer system.
In 2006‑07, the administration costs of the ATO and state revenue offices were around $2.9 billion, or 0.9 per cent of the revenue collected. On the transfer side, the administration costs of Centrelink, which administers the Australian Government income support system, was around $2.5 billion in 2006‑07, or 3.5 per cent of Australian Government payments.
These figures underestimate the total administration costs associated with the tax‑transfer system because they exclude the costs of policy formation, and the costs of other agencies involved in the collection of tax revenue and administration of transfers.
Only two major studies of the costs of complying with the main Australian Government taxes have been conducted in Australia. These were the ATO‑commissioned study by the Australian Taxation Studies Program (ATAX) (Evans et al, 1997) based on 1994‑95 survey data, and an earlier study led by Jeff Pope based on 1990‑91 data (Pope 1994). There has been no aggregate study of compliance costs in Australia since The New Tax System and the Review of Business Taxation reforms were implemented.
The ATAX study found total taxpayer compliance costs in Australia in 1994‑95 to be around 1.4 per cent of GDP, slightly less than the Pope study, which estimated total taxpayer compliance costs to be 2.1 per cent of GDP in 1990‑91 (Table 11.1). The ATAX study found significant differences in compliance costs as a percentage of revenue collected between individuals (4.0 per cent) and businesses (9.4 per cent). It is unclear how representative these estimates might be of existing compliance costs because of subsequent changes in the structure of the tax system and the likelihood that survey respondents may not differentiate the incremental cost of tax compliance from other costs, such as those necessarily associated with running their business.
Table 11.1: Survey estimates of aggregate taxpayer compliance costs in Australia
|% GDP||% Tax revenue||% GDP||% Tax revenue|
- All Australian Government taxes administered by the ATO.
- Personal income tax.
- Employers' PAYE and PPS collections, fringe benefits tax, company income tax and wholesale sales tax.
Source: Evans et al (1997); Pope (1994).
Generally, small businesses face a higher compliance cost burden relative to the amount of tax they pay than large businesses. For example, the ATAX study estimated that small businesses incurred half of tax related business compliance costs, while contributing only around 13 per cent of business tax revenue.
There are no reliable estimates of the costs incurred by transfer recipients to gain and maintain eligibility for payments, or the costs incurred by business in complying with requests for information under the social security law.
There are many examples of complexity in the tax‑transfer system
Previous sections of this paper have identified a range of aspects of the tax‑transfer system that involve considerable complexity, including:
- the large number of taxes levied by governments, many of which raise relatively little revenue, and the lack of consistency in rates, bases and administration of similar taxes levied by different governments (Section 2);
- the many differences in the income tests used to determine eligibility for different transfers and tax concessions (Section 7);
- interactions between the tax and transfer systems that obscure signals about the rewards for working and saving (Section 7);
- the extensive array of tax expenditures, many of which are designed to achieve non‑tax policy objectives through the tax system (Section 2);
- the inconsistencies in the taxation of different assets and entities (Section 8) and the multiplicity of arrangements for pricing Australia's natural resources (Section 2); and
- the difficulties in taxing cross‑border flows of income (Section 8).
<< Previous Page – Section 11.2: Community Concern