Election 2007: Federal-state relations

Roger Wilkins, Citi

The roles and responsibilities of different levels of government in Australia are becoming increasingly unclear. This lack of clarity has allowed ad hoc arrangements to emerge, and encourages sub-optimal policy in vital areas including human services and infrastructure development. Ambiguity makes lines of accountability unclear, has inhibited incentives to produce good policy, has confounded efficient government and undermined the appropriate determination of revenue allocation. How should Australian federalism be reformed? State and federal roles need to be structured so that they are clear, distinct, and work well and the right incentives must be created to support sound policy development.


The key reform is to set in train a process for getting clarity about the roles and responsibilities of the states, territories and the commonwealth based on principle and on rational analysis of what works best, rather than in an ad hoc way that is dominated by political considerations.

There are critical areas of government where the current assignment of roles and responsibilities are either unclear or inappropriate. These are:

  • human services, including health, aged care, child care, disability care, housing and education
  • the planning, provision and regulation of infrastructure, including rail, road, ports, water, and energy
  • the whole area of Aboriginal policy and programmes.

Health is the most prominent example of what happens when there is a perverse assignment of roles and responsibilities. The commonwealth’s interest is to optimise the number of patients going to hospitals because that is a capped programme, significantly funded by the states. The states’ interest is to optimise the number of patients going to doctors and private clinics because they are funded by the commonwealth. The consequent inefficiencies of patients being treated inappropriately are in the millions of dollars, not to mention the costs of inconvenience or worse suffered by patients.


Clear roles and responsibilities for commonwealth and state levels of government are a pre-condition for democratic accountability, for, to put it bluntly, knowing who to blame. Clear roles and responsibilities are a pre-condition for good policy—if you are clearly accountable then the onus is on you to sort out the problems. Clear roles and responsibilities are a pre-condition for efficient government. Perhaps we should add that the assignment of roles and responsibilities also needs to create the right incentives, unlike the situation currently prevailing in the health area. Clear roles and responsibilities are also a pre-condition for sensibly determining the allocation of revenue. There has been too much futile debate about fiscal federalism. Although important, it should be obvious that the first thing we need to know is what expenditure responsibilities governments have, before we can figure out how much money they should raise or get.

There is a massive transfer of money from the common-wealth to the states.

The current situation where the commonwealth raises some 80 per cent of total revenue in Australia but is only responsible for 60 per cent of expenditure is bad for political accountability. There is a massive transfer of money from the commonwealth to the states and territories. This means that the states and territories are not answerable to the electorate for the taxes raised to support their expenditure. And the commonwealth, which raises the taxes, is not accountable for the way the money is spent.


Currently, the roles and responsibilities of federal and state levels of government are unclear. This is a function of two long term trends. First, in a series of decisions the High Court has removed any real restrictions or limits on commonwealth power. More and more power and functions have been centralised. Second, politicians, both state and federal have not been prepared to address this issue in any sort of principled or systematic way. Instead, a sort of ‘ad hoc’ federalism has grown up. Indeed ad hoc federalism reached its apogee with Prime Minister Howard’s Millennium Speech (Howard 2007), which announced an intention, or preparedness, on the part of the commonwealth to intervene directly at any level or in any area of government activity where the commonwealth thought it in the public interest to do so.

This came on the back of recent decisions to intervene in the Murray-Darling, the Davenport Hospital, Aboriginal policy and programmes in the Northern Territory, council amalgamations in Queensland, possibly the regulation and provision of port facilities.

While recent events are probably a caricature of ad hoc federalism, there are real forces at play that are breaking down traditional boundaries and ‘fences’ between the commonwealth and the states. One is the sheer complexity of issues; the way international, national and local aspects are now enmeshed. A paper mill in Tasmania can be seen as a local issue, as a state issue, as a national issue and probably as an international issue. And this is partly a result of the second force: globalisation for economies and increasingly, societies, which is freeing up the movement of capital, goods, and people across political boundaries. What that all means is that we need to re-examine and re-think federalism in Australia.

To some extent this has happened. The reforms of competition and micro-economic policies over the last two decades have effectively broken down many the parochial boundaries that trammelled national markets in, for example, energy, transport, and trade in goods and services. But little in this vein has been done about human services, vast tracts of regulation in areas such as planning and environmental regulation, the regulation of professionals, and consumer credit and transport. Nor have infrastructure and the whole question of Aboriginal policy really been attended to.


Although I am critical of ad hoc federalism, I am not arguing for an approach of national uniformity. There are good and powerful reasons to promote regional and local solutions to address regional and local problems, especially in human resources. There are good and powerful reasons to encourage policy competition and policy innovation, not a ‘one size fits all’ approach. There is also plenty of evidence around the world that people at a local level are wanting more leverage and say over how local services are delivered and local issues resolved.

The High Court has removed limits on commonwealth power.

Federalism which accommodates ‘diversity within unity’ is the type of system that can deal with the dual challenges of globalisation on the one hand and the demand for greater local autonomy on the other. Accordingly ‘co-operative federalism’ is a better approach to the division of labour in a modern federal system, where both the federal government and state governments will have different responsibilities for the same area of policy, say in health, education, or Aboriginal affairs. There could be a national agreement, for example, between the states and the commonwealth government on the outcomes or basic standards to be achieved by states in different policy areas. States are then able to meet those standards or achieve those outcomes in different ways. There may be agreement about certain measures of effectiveness in the health system: re-admission rates or even epidemiological benchmarks for populations. State administrations would then be free to reach these outcomes or standards by funding education/prevention programs, family support programs, and health programs, through public or private providers. And this could vary from place to place, between jurisdictions.

What I call ad-hoc federalism is really an approach to the division of labour between levels of government that looks at a menu of possible relationships between the states and the federal government and then decides which one of those methods to use in each case. Items on the menu include the federal government co-operating with the states; bypassing the states completely; using the states as agents for delivering its detailed policies; and treating state agencies as service providers competing with the private sector. The problem with ad-hoc federalism as it has emerged is not so much this menu of options, but rather that there appears to be no logic or principle or coherent frameworks underlying which option is chosen in a particular situation. The result will be that over time federalism will operate in an increasingly chaotic, unsystematic and ultimately unsustainable way.

The states are not blameless in all this. The states have a tendency to insist on commonwealth funding and assistance in areas where perhaps they should be doing things themselves—for example, most infrastructure projects. And having ‘invited the commonwealth in’, very little thought is given to the longer term shape of that relationship.


One thing that cannot and should not happen is some sort of grand plan or abstract exercise in redefining roles and responsibilities. The politicians are right to resist this sort of abstract theorising. Federalism is not an end in itself, nor can it be divorced as an issue from concrete policy concerns. But that is not to say that principles and a systematic approach have no place. There do appear to be some principles that need to be kept in mind and applied in a presumptive but not slavish way to sorting out problems and issues. These principles do not need to be established from scratch: the Premiers and Chief Ministers have previously enunciated an appropriate set (see D. below).

Federalism is not an end in itself.

The key reform I advocate looks procedural and boring and insubstantial, but many years of looking at and thinking about these things has taught me a key point: that the ‘nerve’ point of many problems is not necessarily some large and dramatic gesture, but rather attention to the detailed dynamics of institutions. So the reform I propose is this:

  1. Establish an agenda for the Council of Australian Governments (the Prime Minister, Premiers and Chief Ministers) (COAG) focused on the area of human services. This proposal is not unlike the National Reform Agenda that has already been agreed to in principle.

  2. Establish a timetable over the next three years for dealing with the substantive areas of policy and programs under that Agenda.

  3. Begin with a report or reports to COAG from experts recommending what governments should do in each of these areas and setting down the costs and benefits that are likely to occur.

  4. Require these reports to deal (explicitly) with the issue of the roles and responsibilities of the commonwealth and the states, applying something like the following four principles communicated by the Premiers and Chief Ministers in Adelaide in 1991.

    • The Australian Nation principle: All governments in Australia recognise the social, political and economic imperatives of nationhood and will work co-operatively to ensure that national issues are resolved in the interests of Australia as a whole;

    • The Subsidiarity principle: Responsibilities for regulation and for allocation of public goods and services should be devolved to the maximum extent possible consistent with the national interest, so that government is accessible and accountable to those affected by its decisions;

    • The Structural Efficiency principle: Increased competitiveness and flexibility of the Australian economy require structural reform in the public sector to complement private sector reform: inefficient commonwealth-state divisions of functions can no longer be tolerated;

    • The Accountability principle: The structure of intergovernmental arrangements should promote democratic accountability and the transparency of government to the electorate (Heads of Government of the States and Territories of Australia, and representatives of Local Government in Australia 1991).

  5. Tie performance and payment: The basis of that payment needs to be clear. If the investment of taxpayer funds works to improve citizens’ health or education then there will almost certainly be growth in the economy and/or costs savings elsewhere in the system. Those costs savings or revenue gains should be equitably shared between the commonwealth and the states and territories. Better still, but less politically practical, would be to reassign taxes so that the commonwealth and states are politically responsible and accountable for raising the taxes they spend.

One thing worth emphasising is that going back to the division of responsibilities implicit in the Australian Constitution before it was reinterpreted by the High Court is not practical. The thought that the commonwealth government could or would relinquish its emerging interest in education, for example, is simply unrealistic and may be undesirable. The issue for the future is how state and federal roles are structured so that they are clear and separate and work well, underpinned and driven by the right incentives. The Principles set out above will not deliver an answer to that question—nor should they. But they do provide a framework for investigating and deliberating about options.


Howard, J. 2007, Address to the Millennium Forum, Four Seasons Hotel, Sydney, 20 August [Online], Available: http://www.pm.gov.au/media/Speech/2007/Speech24507.cfm [2007, Sep 16].

Heads of Government of the States and Territories of Australia, and representatives of Local Government in Australia 1991, Leaders’ Forum Communiqué, Adelaide, 21–22 [Online], Available: http://www.premiers.qld.gov.au/policy/intergovt/coagmincncl/Communiques/Leaders_Forums [2007, Sep 16].

Roger Wilkins was Director General of the NSW Cabinet Office for fifteen years and Director General of the NSW Ministry of the Arts for five years. Since November 2006 he has been Head of Government and Public Sector Group with Citi. Roger Wilkins is also on the Board of the International Forum of Federations.

View other articles for Election 2007: