Skip to main content Skip to navigation
Australian Government - GST Distribution Review

Chapter 9 - Efficiency and tax reform

A key issue for the Review has been whether the pursuit of HFE leads to unacceptable trade-offs with efficiency, or discourages tax reform. For a variety of reasons, many submissions (including those from New South Wales, Victoria, Queensland and Western Australia) have argued that the current system of HFE impedes tax reform and creates other efficiency reducing distortions. On the other hand, many other submissions (including those from South Australia, Tasmania, the Australian Capital Territory and the Northern Territory) have argued that HFE is generally efficiency enhancing.

The Panel’s interim reports thoroughly examined concerns about the relationship between HFE, efficiency and tax reform.1 Having considered further comments on these issues in subsequent submissions, the Panel has formed the view that the current system can and does create perverse incentives in theory, but that there is little evidence of those incentives having any effect in the real world. In particular, there is no evidence that HFE acts as a material disincentive to State tax reform.

While there may be some merit in addressing perverse incentives on principle alone, after exploring the alternatives, the Panel has concluded that they cannot be meaningfully reduced without significant reductions in equalisation outcomes. As there is little evidence of real world efficiency losses, the Panel is not convinced that this would be a worthwhile trade-off.

On tax reform, the Panel believes that it is vital for all levels of government to pursue a tax system that favours broadly based taxes with fewer exemptions over narrow and distortionary transaction based taxes. Ideally, this would occur on a multilateral basis — amongst the States and including the Commonwealth — but it would not be a bad thing if some States chose to take a leadership role.

In relation to each of the supplementary Terms of Reference specifically:

  • There is no clear evidence that the current system of HFE is impeding State tax reform. If it ever became apparent that possible changes in GST shares were impeding tax reforms, specific and temporary adjustments to the GST distribution should be made in that context, rather than changing HFE.
  • While the Panel’s Second Interim Report canvassed the possibility of GST grants being used to reinforce incentives to undertake agreed upon reforms, the Panel has concluded that the GST distribution should not be used to compel or encourage States to change policies — HFE should be policy neutral and GST should remain untied and freely available.
  • A national agreement is needed in relation to resource taxation. Chapter 8 of this Report describes how such a deal might be achieved. HFE is not a significant barrier.

9.1 Relationship between HFE and efficiency

Does HFE encourage States to manipulate tax or spending policies to maximise their GST share?

A State is able to increase its GST share by increasing its policy effort (that is, raising higher taxes or spending more on services) in areas where the CGC assesses that it has below average capacity and decreasing its policy effort in areas where the CGC assesses it has above average capacity.

The median magnitude of these incentives is less than 1 per cent of the change in State spending/revenue and less than 5 per cent in all cases except mining revenue. Intuitively, this seems too small to have any decisive impact on State decisions, particularly in the context of other drivers, such as public preferences for services and the gross incidence effects of tax mix switches. Empirically, there is no obvious correlation between GST incentives and a State’s policy effort. Indeed the Panel has not been presented with even anecdotal evidence of GST incentives playing into State decisions, and there are many recent examples of States acting in spite of them.

Although there are some issues to resolve with the mining revenue assessment, the Panel is not convinced that the concern that HFE encourages States to manipulate policies to maximise their GST share provides grounds to change the system.

Does HFE discourage tax reform?

While the GST share effects that result from a single State changing its policy are relatively small, their magnitude increases when several States act simultaneously.

Major tax reform is likely to require a tax mix switch. There are different ways to approach a tax mix switch, including on an own-source revenue neutral, or a net budget neutral, basis. If a State seeks to raise the same amount of revenue from both the new tax and the old tax, changes in GST can have a significant impact on State budgets. On the other hand, if States were to maintain the same tax rate (and coverage) relative to other States, there would be no net budget impact, as States that received less GST would receive more own source revenue and vice versa.

Regardless of how tax reform is pursued, under the current HFE arrangements, there will always be gains and losses in GST shares. While these gains and losses will be equal across the system, they will affect States differently, with some States getting a reduction in GST share and others an increase. In practice, where a State’s GST share is reduced, that State would generally view itself as a ‘loser’, notwithstanding that it could avoid the budget impact by collecting more from the replacement taxes.

Nevertheless, a number of factors other than changes in GST shares drive State tax policy decisions, in particular, the political constraints associated with large changes in the gross incidence of taxation. The significance of changes in GST shares to State tax policy decisions is probably diminished further by the delay in changes to GST being felt — because of the lag, any change in GST will not occur until several years after a tax policy change has been implemented.2

Examining the empirical evidence, States do not necessarily appear to act in accordance with the apparent GST share incentives. Ultimately, there is no hard evidence on whether GST share effects influence State tax reform decisions. The Panel doubts that GST share effects are a very powerful factor when States are considering tax reform.

Does HFE encourage higher taxes?

Another concern put to the Panel is that HFE does not take into account the effect of a State’s tax rates on the size of its tax base. If a tax base is relatively elastic (meaning that activity is relatively sensitive to price changes), then a change in the tax rate could have a significant impact on a State’s GST share (unless tax policy changes are undertaken multilaterally).

For example, the Australian Capital Territory raised the concern that its moves to switch from stamp duty to a more broadly based land tax would cause it to lose GST by increasing the volume of housing transactions.3 Similarly, New South Wales argued that, although cutting its insurance tax rates would increase its GST share by lowering the average tax rate in a tax where it is assessed to have above average capacity, it would ultimately lose GST because more people would take out insurance. If more people in New South Wales take out insurance then it would be assessed to have an even higher capacity than it currently does.4

Adjustments were previously made to some revenue assessments to take into account elasticity effects, but this was discontinued because of the lack of a reliable measure. An attempt by researchers in 2002 to calculate the size of elasticity effects in respect of land tax found that if a single State increased its rates it was likely to receive a GST increase of between two and three cents for each additional dollar of land tax. New South Wales argues that modelling to support the Australia’s Future Tax System (AFTS) review, suggests the elasticity of State tax bases is much higher, and therefore the effect on GST shares is much larger, particularly for inefficient taxes like insurance duty.5

Elasticity effects could have a relatively significant impact on a State’s GST share if it moved to abolish an inefficient tax unilaterally. However, in circumstances where a State’s tax base did observably change significantly following a major tax policy change, the CGC may be more amenable to introducing elasticity adjustments.

As with other theoretical incentives, there is no evidence that HFE has affected State decisions. Indeed it seems unlikely that States would increase their taxes so that they could receive more GST, given the amount would be uncertain, small relative to the overall effect on revenue, and at the expense of a large tax increase. However, at the margin, HFE may be a disincentive to lower tax rates.

While the Panel does not believe that these effects have a significant influence on State policy, failure to account for elasticity effects may lead to less accurate equalisation outcomes. However, accounting for elasticity effects would make the system considerably more complex.

Does HFE discourage policies to promote economic development?

If a State’s share of the tax base expands its GST share is reduced, so that it effectively only receives a per capita share of additional tax revenue (plus or minus the effect of any above or below average tax rates). For example, if employment increases in Queensland it will raise more payroll tax revenue, but if employment in other States remains steady or falls, Queensland’s share of the payroll tax base will also increase. However, Queensland’s GST share will be reduced so that it effectively only receives its population share of the increase, adjusted for its below- or above-average tax rates. This represents a significant reduction in the fiscal incentives that a State has to implement policies designed to increase its per capita tax base.

However, increases in a State’s tax base are generally the result of economic growth, and a State has many reasons to pursue economic development beyond the potential fiscal gains. Nonetheless, the Panel has recognised that some of the costs associated with pursuing mining related activity may not be adequately recognised in the HFE system. Ensuring adequate recognition of these industry support costs will remove a potential discouragement from promoting mining related activity.6

Once again, there is no clear evidence of any instance when GST share effects have changed a State’s decisions.

Does HFE discourage efficient service delivery?

Some States are concerned that, because HFE recognises the prevailing cost of delivering services, States are discouraged from pursuing efficiency and cost reductions.

In the case of cost savings, clearly the factors motivating States to pursue efficiency are primarily fiscal, and any significant reduction in those fiscal incentives would be of concern. On the other hand, if a State is able to reduce the amount it spends, say by more efficiently servicing remote communities, that State is able to keep the entire saving, bar some minor effects that emerge from the impact on average policy.7

It is true that a State would not receive fiscal benefits beyond its population share from addressing the underlying drivers of cost factors (that is, not just how much it costs to provide services to remote communities, but the numbers of remote communities). This is also true in relation to use factors, such as the number of elderly and school aged children. This is problematic, but only to the extent that such factors are within a State’s control. The CGC deliberately seeks to avoid equalising based on factors that a State can control. New South Wales provides the example of private versus public school enrolments as a factor that States can influence.8 If States showed any inclination to move towards a voucher-based model of education funding, there may be some grounds to modify that assessment. However, States have not, and the potential GST effects have not been the defining factor.

In general, it is much less clear that States have any control over the disability factors included in the CGC’s assessments — for example, States are not realistically able to control the number of elderly people within their State. Nor would policies aimed at reducing the number of elderly be acceptable. Similarly, when it comes to decisions about whether or not people should be more or less dispersed there are many non-fiscal reasons why States do not adopt a policy of ‘forcing’ a centralised settlement pattern.

Does HFE discourage efficient migration?

Some submissions express the concern that HFE discourages efficient migration, because in the absence of HFE more people would live in Western Australia to take advantage of the better fiscal deal it would be able to provide. They argue that the extra incentive is necessary to counteract other government policies (such as stamp duty) which impede mobility.

However, individuals and businesses may be inclined to locate in States with stronger economic growth anyway and it is difficult for government to determine what the appropriate settlement pattern should be. The change in the outlook for mining over the course of this Review indicates the difficulties involved in making such predictions.

Even if the ideal settlement pattern were known, it is not clear that modifying HFE would be the best way to achieve it as it would involve paying for additional services (or lower taxes) for everyone in particular States, not just those who move. There are also some specific concerns in relation to location costs.9

Does HFE create grant dependency?

Separate from the specific effects of changes in the GST distribution on State incentives, an underlying theme in some submissions is that recipient States become dependent on grant funding and are less likely to pursue policies to improve their circumstances.

Garnaut and FitzGerald10 argued that having a larger proportion of public sector workers in recipient States creates distortions in the local political economy. The theory is that States with a larger proportion of public sector workers will be more inclined to prefer social and environmental objectives to economic ones. Garnaut and FitzGerald note that the scale of this effect is difficult to quantify, but suggest that it may be significant.

There seems to be little firm evidence to support these concerns. As noted in Chapter 5, State shares of Commonwealth support are much closer together when all Commonwealth payments are considered and State performance on economic and social indicators is often much closer than differences in fiscal capacity. Differences in fiscal capacity (or other indicators) can largely be traced to geographic, demographic and economic factors outside a State’s control. While State governments sometimes make poor decisions, this is surely true of both donor and recipient States.

Does HFE lead to excessive provision of public services?

In their 2002 review of the distribution of the GST and Commonwealth payments, Garnaut and FitzGerald argued that intergovernmental transfers like the GST tend to ‘stick’ with governments and are used to fund public services rather than being allocated back to citizens in tax cuts resulting in excessively large public sectors in recipient States. As explained in Chapter 1, this is known as the flypaper effect.

The obvious explanation for this is that grants are generally given to State and local governments with the explicit purpose of being used to fund public services, particularly in the Australian context where the service responsibilities of State and local governments generally exceed their taxation powers. Moreover, as the Commonwealth Treasury noted in its submission to the Review, it is not true that recipient States as a whole have larger public sectors. Instead, those States that spend more are the ones that are assessed to have above average spending needs.11

Finding 9.1

The current system creates perverse theoretical incentives in some instances, but there is little evidence that they have any effect in the real world. In particular, there is no evidence that HFE acts as a material disincentive to State tax reform.

There may be some merit in addressing perverse incentives on principle alone. However, after exploring the alternatives, the Panel has concluded that they cannot be meaningfully reduced without significant reductions in equalisation outcomes. As there is little evidence of efficiency losses in practice, the Panel is not convinced that this would be a worthwhile trade-off.

One area where there may be merit in further investigation by the CGC is in relation to the impact of tax rates on the size of State tax bases.

9.2 Tax reform

What might tax reform look like?

Findings on State taxes in the AFTS report and the outcomes of State tax reviews are useful starting points for a discussion between States and the Commonwealth on identifying a preferred future tax mix and establishing a preferred tax reform path.

AFTS recommended replacing less efficient State taxes with more efficient ones. A broader based land tax, for example, could replace the existing land tax and stamp duty on conveyances. Motor taxes could be replaced with an improved road user charging system. Insurance taxes could be replaced in a number of ways, including an improved payroll tax, and the payroll tax itself could be replaced by a business cash flow tax.

Whatever the final shape and direction of reform, the recommended general trend can be summarised as:

  • placing more reliance on a few main taxes and less on smaller ‘nuisance’ taxes
  • replacing less efficient taxes with more efficient ones (which usually means replacing high impact, narrow-based taxes, with broader-based, lower-rate ones)
  • raising more revenue from less mobile tax bases and raising less revenue from the more mobile tax bases.12

While AFTS was precluded from examining the rate and base of the GST, it is worth pointing out that the business cash flow tax has much the same economic incidence as a GST with a broader base. Similarly, the payroll tax and the GST have much the same economic incidence, with overlapping, but not identical bases.

How should tax reform be pursued?

Of the States that raised State tax reform in submissions, all agree that the ideal way to achieve reform is through a multilateral negotiation or discussion that includes the Commonwealth. Many States also identify the approach taken in implementing the A New Tax System (ANTS) reforms as a model for future State tax reform.

AFTS recommendation 119 also urged the use of intergovernmental agreements between the Commonwealth and State Governments to coordinate State tax reform by providing the States with revenue stability and facilitating good policy outcomes.

Any effective intergovernmental agreement on State tax reform would identify what taxes would be abolished, amended, or reformed, the timeline for achieving the reforms, and the means to address any GST share effects — if they are considered important. While State tax reform does not strictly require Commonwealth involvement, the Panel would encourage the Commonwealth to be actively engaged in any proposed reforms, as State tax reform will be important in positioning the Federation to respond to future challenges. The Commonwealth’s role could be as a neutral facilitator in deliberations among States on the direction and pace of reform. On the other hand, States that wish to pursue tax reform independently should not feel constrained by the need for a multilateral consensus.

Dealing with any GST effects

While the Panel is comfortable that HFE is not preventing State tax reform, it is clear that tax reform would result in changes in States’ GST shares. Various options for addressing these effects in the system as a whole have been considered, including equalisation to a minimum standard, an external standard, or using broader indicators. The Panel has opted not to pursue these options at present as any potential efficiency gains are minimal and are outweighed by larges losses in equity.

If changes in GST shares do prove a barrier to some future tax reform, they should be addressed in the context of the specific tax changes that are being proposed. In this regard, the Panel’s second interim report discussed the possibility of the Commonwealth providing compensation for GST share effects.13 As long as the reform process led to quantifiable fiscal benefits for the Commonwealth, compensation would not breach the Commonwealth’s injunction that it will not fund State tax reform.

Another option considered was using any fiscal dividend from reform to adjust the GST distribution so that in the case of major multilateral changes, GST share effects did not occur. Like other options to break the link between State policy and the GST distribution, this option may raise equity concerns, but in a much more narrow context than an overarching change to the distribution of GST. Moreover, these equity concerns may be outweighed by the efficiency gains if compensation for GST effects allowed tax reform to proceed when it would otherwise have failed.

Reinforcing agreed reform

The Panel’s second interim report outlined the view that the GST distribution should remain untied and freely available, and not be used to provide positive incentives for tax reform. Nevertheless, it also canvassed various ways that the GST distribution might be used to reinforce agreed upon commitments. However, such a change to the GST distribution would be dependent on State support and there was no support for these options in subsequent submissions.

Finding 9.2

The Panel is convinced that it is vital for all levels of government to pursue a tax system that favours broadly based taxes with fewer exemptions over narrow and distortionary transaction based taxes. Ideally, this would occur on a multilateral basis — amongst the States and including the Commonwealth — but it would not be a bad thing if some States chose to take a leadership role.

In relation to each of the supplementary Terms of Reference specifically:

  • There is no clear evidence that the current system of HFE is impeding State tax reform. If it ever became apparent that possible changes in GST shares were impeding tax reforms, specific and temporary adjustments should be made to the GST distribution in that context, rather than changing HFE.
  • While our second interim report canvassed the possibility of GST grants being used to reinforce incentives to undertake agreed upon reforms, the Panel has concluded that the GST distribution should not be used to compel or encourage States to change policies — HFE should be policy neutral and GST should remain untied and freely available.
  • A national agreement is needed in relation to resource taxation. Chapter 8 of this Report describes how such a deal might be reached. HFE is not a barrier.

1 GST Distribution Review, Interim Report, March 2012, Chapter 4, and GST Distribution Review, Second Interim Report, June 2012, Chapter 3.

2 For an explanation of the lag, see GST Distribution Review, Interim Report, March 2012, page 28.

3 Australian Capital Territory supplementary submission to the GST Distribution Review, March 2012, page 2.

4 New South Wales third submission to the GST Distribution Review, August 2012, pages 59-61.

5 New South Wales also argues that it is only assessed as having above average capacity in relation to insurance duty because it has a below average tax rate. However, it is not clear that this explanation is correct, given New South Wales also imposes a fire services levy on insurance premiums, which moves its overall tax rate close to the national average.

6 See Chapter 7.

7 See Commonwealth Grants Commission, Report on GST revenue sharing relativities, 2012 Update, Box 2-1, page 31.

8 New South Wales submission to the GST Distribution Review, November 2011, page 29.

9 See Chapter 6, Section 6.4.

10 Garnaut, R and FitzGerald, V, Review of Commonwealth-State Funding, Final Report, Melbourne, 2002, pages 145-6.

11 Commonwealth Treasury submission to the GST Distribution Review, October 2011, page 35.

12 Australia’s Future Tax System, Report to the Treasurer, Part 1, Overview.

13 See GST Distribution Review, Second Interim Report, June 2012, Chapter 5, Section 5.4.