Australia's Future Tax System

The Future of State Revenue

Speech by Dr Ken Henry
Chair – Australia's Future Tax System Review Panel and Secretary to the Treasury

2009 Commissioners' Conference, Sydney
27 March 2009

I would like to start by thanking the organisers for the invitation to talk to you about state taxes and their administration.

I have the chance to meet with my Heads of Treasuries colleagues a number of times each year – most recently, only a couple of days ago in Canberra. The HoTs, as it is called, provides a valuable forum for exchanging ideas, including about state and territory taxes. So it is great also to have the opportunity to talk with those deeply involved on the administration side. After all, the state revenue offices are responsible for the collection of over $45 billion in taxes each year, playing an extremely important role in Australia's tax-transfer system.

Australia's Future Tax System

Much of what I have to say today will be in the context of the opportunities and challenges that the Australia's Future Tax System review presents for state taxes and federal financial relations.

The tax review panel is part way through a review that has been described as the most comprehensive review of the Australian tax-transfer system for at least the last 50 years. The terms of reference set an objective for the review of creating a tax-transfer structure that will position Australia to deal with its demographic, social, economic and environmental challenges, and enhance Australia's economic, social and environmental wellbeing.

With that objective, it was crucial that the taxes and transfers administered by the states and territories be included in the terms of reference for the review. Their inclusion has given the review panel the scope for a comprehensive review of all taxes and transfers across all three levels of government in Australia.

It means that the Panel can assess not only how different taxes and transfers rate against the standard policy assessment criteria, but also how appropriate it is for the different taxes and transfers to be assigned to a particular level of government.

In the Panel's view, it is not possible to assess the current structure of federal, state and local taxes and transfers without also considering the appropriate financial relationships between Australia's three levels of government.

In Australia's history, federal financial relations, primarily the financial relationship between the Commonwealth and the states, have changed in an ad hoc way, with reforms often designed to support other policy objectives or to address a crisis at a particular point in time.

In contrast, our review is explicitly forward looking and comprehensive. And it is certainly our intention that its recommendations not appear ad hoc. We have the opportunity to think about how governments at all levels might, even should, operate in the future and how the tax-transfer system might best be designed to support this.

While the review is comprehensive, there are some issues which the Panel will not be considering in depth. For example, given the recent commitment in the Intergovernmental Agreement on Federal Financial Relations and the Commonwealth Grants Commission's current methodology review, the Panel considers that reviewing the principle of horizontal fiscal equalisation is beyond its scope.

Today I want to talk about some of the challenges that will be faced in the future and how they will impact on state finances. And then I want to discuss how these challenges influence the way we think about the design of state taxes in terms of policy, administration and revenue distribution before looking at a few examples of how our thinking might be applied in practice.

Challenges and opportunities for reform

In our consultation paper released in December last year, the review panel identified a number of broad issues to frame consideration of Australia's future tax-transfer system. Three of those issues – increasing globalisation, demographic change and the role of technology – are particularly relevant to state finances.

Just as economic activity expanded beyond local communities in the distant past, recent decades have seen the increasing integration of economic activities across countries. Despite concerns from some about its impact, we can expect the process of globalisation to continue. Globalisation means that the things governments tax are becoming increasingly mobile. This has implications for tax system design.

For example, while the real employment consequences of the global financial crisis are obviously highly significant, and will have a substantial impact on global migration flows in the short term, the long-term trend is that increasing numbers of highly skilled individuals are operating in a global labour market. This international mobility will impact on the way that labour is taxed at a national level, let alone at a sub-national level.

But there are some inherently immobile tax bases, such as land. The importance of taxing these bases effectively is likely to increase in the future.

The second challenge, the challenge of demographic change, has been highlighted by the Intergenerational Reports, and by the Productivity Commission in its complementary report 'Economic Implications of an Ageing Australia'.

Demographic change will affect different tax bases and types of government expenditure in different ways. Governments at the federal, state and local level will face fiscal pressures at different times. And these pressures will influence the financial relationship between the different levels of government.

The third challenge highly relevant to state finance concerns how well we are able to use and adapt to new forms of technology. Emerging technologies have the potential to redefine the way we design and administer the tax-transfer system, with significant implications for both compliance and operating costs.

One of the best illustrations of this is in the area of road pricing, where new technologies can allow for more efficient charging for road use, leading to reductions in congestion and improvements in traffic flows. For example, in Singapore the road pricing system charges each car a price to use the busiest roads at the busiest times of the day. The price varies according to the day and time and users can have complete information about the cost of their trip before it is undertaken.

The geography of Singapore is a bit different from most Australian cities. But the fact that the Singaporeans have been pricing road access since the mid 1970s, and have been doing so electronically for the last decade, shows that variable road pricing is not science fiction. I will say a little more about this later.

Technology can also improve the timeliness and reliability of information flows and the capacity for people to access and understand information about the tax-transfer system. This not only creates the potential to reduce administration and compliance costs, but gives us scope to rethink the way that each level of government may best contribute to the administration of taxes and transfers.

Views from submissions

So there are both challenges and opportunities for governments at all levels to improve the way they levy taxes and provide transfers and government services. And there is much that can be improved upon.

As part of the review, we have invited the public to make submissions on their views about the future challenges and current problems facing the tax-transfer system. We have received over 1,000 submissions to date from people and organisations across the entire community – an extremely valuable resource.

While taxes in general are – unsurprisingly – not popular with the community, the submissions reveal that state taxes are among the least popular.

There were many submissions which called for the abolition of particular state taxes, noting specific inefficiencies, inequities and complexities with these taxes. Some submissions went as far as to advocate the abolition of all state taxes.

On the positive side, there were submissions noting that the states have access to some potentially good tax bases and recommending improvements in the form of base broadening.

Many submissions also raised the issue of the complexity created for businesses operating across state borders from each state and territory government having its own administration responsible for the collection of taxes.

To paraphrase one submission, while legislation has been harmonised for some taxes, there are still areas where there is a significant difference in the application of taxes. This adds a burden to those businesses that operate across state borders and makes it harder for them to expand. Consistency in the design of state taxes would help businesses operating across state borders to reduce compliance costs and better understand their obligations.

Taxation in a federal system

Given the challenges and opportunities that we face and the concerns that have been expressed about the operation of state taxes, how should taxes be levied across the federation in the future? First of all, we should not forget that taxes need to rate well against the principles of fairness, efficiency, simplicity, sustainability and coherence. And then we should consider how taxes meeting these criteria can best be applied in the federal system.

We can think of three dimensions to this assignment exercise – the level of government responsible for the design of the tax; the level of government responsible for administration and collection of the tax; and the level of government that receives the revenue raised by the tax.

The design dimension is centred on the question of who is responsible for setting the rate and base for the tax. The theory of tax assignment developed by Richard Musgrave and others suggests that in a federal system, lower level jurisdictions should avoid using taxes for the purposes of income redistribution and macroeconomic stabilisation. These functions are more appropriately assigned to the national level. And since the mobility of tax bases impacts on how effectively governments can levy taxes, in general lower level jurisdictions should avoid tax bases with high inter-jurisdictional mobility.

It is normally the case that the rate and base of a tax is set by the one government. But it is also possible that the rate and base be set by different levels of government. For example, one can conceive a set of arrangements in which states levy different rates on a nationally consistent tax base.

Such an approach could avoid problems with tax base erosion from interstate competition, and make it easier for businesses to understand and comply with their obligations, while still providing the states with a policy lever to respond to jurisdiction-specific preferences.

The administrative dimension is a question of which level of government is more appropriately responsible for administering and collecting the tax. In many cases, the answer to this question will be dependent on the policy dimension – the level of government that sets the policy will usually be better placed to determine how that policy should be implemented.

However, if several governments at a sub-national level levy taxes which are very similar, then there may be economies of scale from a centralised administration. The realisation of these economies becomes more feasible as better flows of information between taxpayers and administrators.

Thus, it seems to me to be a genuinely open question whether, in years to come, Australia should persist with state and territory revenue offices.

The third dimension concerns how the revenue from taxes is distributed between levels of government. It is usually the case that whoever controls the policy and administration will also receive the revenue – and it is important that governments have some capacity to alter revenue consistent with their marginal expenditure choices.

But it is also usually the case in federal systems that there is an imbalance between the revenue that each level of government raises and its expenditure requirements. For some taxes, therefore, part or all of the revenue may be given to another level of government. Then there is the question of how this revenue is distributed among governments at the same level and with what conditions.

There are trade-offs to be made in this three-dimensional assignment task. The more the policy and administration of the tax system is centralised at the national level, the greater the opportunity to develop a less complex and more efficient tax system.

However, centralisation obviously also means that sub-national governments have a greater reliance on revenue from the national government. And this may influence their spending decisions.

The arrangements supporting the GST provide one answer to the three-dimensional assignment question, reflecting a serious consideration of these trade-offs. The policy for the GST is legislated by the Commonwealth. Under our constitution, there is no alternative. But the Commonwealth is also responsible for the administration and collection of the tax, while the states receive the revenue from the GST, fund its administration, and have to agree to any legislative changes to the GST base. These arrangements mean that the tax base and rate is consistent across all states and both levels of government have an interest in both the tax base and how well the tax is administered.

Some submissions to the review noted that a similar model could be applied to payroll tax, or some other form of a broad-based labour tax. In these submissions, the base would be completely harmonised across the states, the administration and collection of the tax would be carried out by the Commonwealth, while the states would receive the revenue and – a key difference from the GST – maintain the ability to set different rates.

Some of the submissions acknowledged the benefits of the recent process of payroll tax harmonisation undertaken in recent years. But they argued the case for taking this next step. Without explicitly endorsing the proposal at this time, I would note that it would also provide the opportunity to link payroll tax to the PAYG regime, and to broaden the base of one of our potentially least distorting taxes.

Taxing resources in a federal system

A more complex area is resource taxation. I have noted that the taxation of immobile bases will become an increasingly important matter in the context of increasing globalisation. One of those bases is our non-renewable resources.

Referring to our non-renewable resources as a tax base is rather crude. While it is convenient to refer to the 'taxation' of resources, the royalties and other charges imposed by the Commonwealth and the states represent a return to Australians for assigning certain rights to appropriate – exploit, if you prefer – those resources. They represent a disposal price. It is always interesting to hear people refer to the various royalty regimes as secondary taxation when their very nature indicates a primacy.

The three dimensions of taxation in a federal system are also relevant here: Who should design the tax? Who should be the administrator? Who should get the money?

The system for ensuring that the Australian community receives an appropriate return on the disposal of its non-renewable resources is multi-faceted to say the least. All states have resource charging arrangements that vary in type, the resource upon which they are imposed and the rate. The Commonwealth is a big player in resource charging for our offshore petroleum resources – again with a variety of arrangements. Company tax also plays a role in ensuring a community return on the disposal of its non-renewable assets.

Given this starting point, it is worth asking if there could be any gains from the Commonwealth exercising a greater role. First, there is the 'David and Goliath' argument that states may be in a weak bargaining position when negotiating with the developer of the resource. Notwithstanding the fact that the location of a resource is fixed, resource firms will typically present themselves to a state as having to make decisions about competing projects, perhaps in other states. While it is one thing to compete internationally for investment, bargaining between states does not benefit Australia overall.

Second, a single comprehensive regime at the Commonwealth level could be seen as being less likely to be subject to change, therefore reducing sovereign risk. This, at least, is the experience with the Commonwealth's Petroleum Resource Rent Tax, which has not been subject to significant change since its introduction.

Third, there are tax design issues. It is generally accepted that a profits-based royalty arrangement, particularly one based on capturing a share of the economic rent (or above normal profits) of a resource project, is more economically efficient than an output-based royalty arrangement. The petroleum resource rent tax is a type of profits-based royalty designed to capture economic rents. Submissions to the review suggest some private sector support for profits-based royalties.

In contrast, the states have favoured output-based royalty arrangements, whether fixed rate or ad valorem, in part because of their more predictable revenue stream and low administration and compliance costs. An exception to this is the Northern Territory, which operates a profits-based royalty regime. Interestingly, it is the only state or territory where resource charging arrangements are administered by the revenue agency as opposed to the mining or resource department.

The issue of unifying resource charging arrangements, or at least a common regime, has been tested in recent times in a review conducted by the relevant ministerial council on resources. The outcomes of the review suggest that the states are comfortable with their heavy reliance on output-based royalties. That there was not a lot of support for profits-based royalty arrangements suggests unease about the potential lack of revenue flows in the early years of a resource project and the risk of lower revenue flows because of lower profitability.

The attraction of output-based royalties for the states, notwithstanding their potentially negative impact on low profit or marginal projects, is that stable revenue flows better match the states' expenditure responsibilities. Even so, resources are finite. So even output-based royalties won't be stable in perpetuity. This explains the creation, by countries with large resource endowments such as Norway, of sovereign wealth funds that accumulate revenue from resource charging.

If the most efficient charge is a profits-based royalty, but the states would prefer a less volatile revenue stream we might have an argument in favour of the Commonwealth taking the lead in policy. There may also be an argument in favour of Commonwealth administration. But the third dimension of the assignment problem – who gets the revenue – becomes really tricky.

In the examples thus far I have pointed to factors, or arguments often raised, for a greater role for the Commonwealth in policy design and administration. But such arguments don't support the greater centralisation of all taxes and charges. Where bases are relatively immobile and the collection of revenue is dependent on localised information, then lower levels of government will usually have a comparative advantage in policy and administration.

The example of road pricing is again instructive. It would be impractical for a national government in a very large country with diverse cities to try to design road pricing policies across all major urban areas. Pricing arrangements would need to be city-specific, because of the variability in the risk and character of spill-over effects, including impacts on public transport systems. On the other hand, new and emerging technologies increasingly support the feasibility of one administration agency. And, as in the case of non-renewable resource taxation, the most interesting issues relate to the third dimension of the assignment problem; that is, to the question of revenue flows.

But imagine this: a national system in which the driver is charged – on a per-kilometre basis – for access to the road network, with the charge depending upon the incremental damage to the road surface, quality of the road (whether dirt or sealed), and so on. Imagine a system that ensured that a local council in the west of NSW, responsible for the maintenance of a timber bridge, received a small payment every time a car travelled over the bridge, a larger payment from a two tonne truck driver, and an even larger payment from an articulated lorry driver.

Improving the federal structure of the tax-transfer system is a key issue for the review. And it is one in which we all have a stake since it is an important determinant of governments' ability to enhance economic performance and the broader wellbeing of Australians. I look forward to discussing these issues with you.