Federalism - Australia and New Zealand School of Government - 11 September 2008
In other articles I have drawn attention to four different models or ways in which “cooperative” federalism could develop.
The first model I called the “partnership” model. This is based on a transfer of funds from the Commonwealth to the States under Section 96 of the Constitution. The purposes or objectives of the expenditure are mutually agreed between the Commonwealth and the States. And it may well involve agreement to commit State funds for the same purpose.In effect, this is the sort of model the Rudd Government has been developing through COAG.
The second model is the “agent” or “coercive” model of federalism. Like the partnership model, money is transferred from the Commonwealth to the States under Section 96. The essential machinery is the same. But in this case there is no mutual agreement. There is much more a coercive use of funds. “You can only have these funds if you agree to put such and such in your curriculum … and you agree to match the funds.”
The third model is the “contracting model” of federalism.This is where the Commonwealth essentially treats the States in the same way as other private sector players in a market. The States have to compete for Commonwealth funds, and contract to deliver services.This was a model the Howard government appeared to be moving toward with training colleges and some environmental and water projects. Amore extreme form of this would be a “public choice model” of federalism.Where the Commonwealth provides money or vouchers or entitlements directly to citizens and effectively allows citizens to decide where the money should go. The State service deliverers have to compete with the private sector to attract citizens and funds.
The final model is the “coordinate model” of federalism. Effectively this is not a form of cooperative arrangement at all. No, it is a model of federalism where States are adequately funded through their own revenue to undertake their functions and require no Commonwealth funds. In an odd and, perhaps, ironic way this is where the “GST Deal” leads. The GST Agreement, you will recall, is an agreement whereby the States are given access to the entire proceeds of the Goods and Services Tax. In return they agreed to return certain inefficient taxes. But the important thing is that the States have access to a growth tax. Over time, perhaps a very long time, that could mean that the Commonwealth loses financial leverage over the States. There are some big “if’s” in that. “If … the Agreement holds … If the Commonwealth does not begin to put new conditions on payment etc.”
Anyway, it is worth keeping these different models or scenarios in mind as I speak. And remembering that from an institutional point of view all of these scenarios are possible futures under the Australian Constitution. In other words the Constitution leaves the shape and fundamental nature of federation ‘radically under determined’. (At least as it has so far been interpreted.)
There has been a lot of talk in Australia recently about a new era of "cooperative federalism". It is a phrase that keeps cropping up. After the last few meetings of Premiers and Chief Ministers with the Prime Minister, the term was frequently used. With wall-to-wall Labor governments in Australia, "cooperative federalism" seems to be the way leaders want to characterize the way forward for the Australian federal system.
I want to begin by making a few observations about "cooperative federalism". I do not want you to misinterpret what I am going to say because, in many respects, I am an advocate of "cooperative federalism". But my prefatory observations are aimed at putting "cooperative federalism" in perspective, and as an antidote to some of the rather "purple passages" that have been pronounced on Australian federalism.
First, "cooperative federalism" is actually a technical term. It describes a system of federalism not where there is necessarily "cooperation" in the ordinary English sense of the word, but rather, where there needs to be cooperation between levels of government to get things done in the system.
"Cooperative federalism" is not to be contrasted with "uncooperative federalism", but with systems of "coordinate federalism" where different levels of government can get on with the business of government without necessarily co-opting the other levels of government. The US or Canada are examples of "coordinate" systems of federalism. Germany is an example of a "cooperative" system of federalism.
Let me sketch out for you in a little more detail the contrasts between these two different ways of doing federalism.
Take health for example. The US has a rather special public health system. One might say rather marginal because of the overwhelming dominance of the private sector. But in Canada, it is more or less a matter for the provinces. The provinces, with a few exceptions relating to indigenous peoples for example, are responsible for Canada's health care system. True it is that the provinces have tried to get the Federal Government to provide greater funding assistance. It is also true that issues around the NAFf A have begun to impact on the way health is delivered in Canada, with US managed care corporations trying to infiltrate Canada. And it is also true that, as in Australia, there are manpower and training issues that tend to bring the Federal Government into play. But, the centre of gravity, in terms of health and the overwhelming responsibility for health, lies with the provinces; and is not really split between the two levels of Government.
In Germany, what you find is a real division of labour between the Federal Government and the Länder. The Federal Government does not deliver health services but does provide the regulatory framework and policy settings within which the Länder and the Gemeinde and the private sector provide health care. This includes issues such as insurance, quality control, some funding, national priorities in terms of health. Responsibility really is split but it tends to be a division of labour between the setting of regulatory and policy frameworks (Federal) and the deliver of services and provision of infrastructure (Länder /Gemeinde).
The second thing I want to say about "cooperative federalism" is that it is not a case of everyone being responsible for everything. In Western systems of cooperative federalism, it is still important to define the roles and responsibilities of the different tiers of government. So, for example, as we have noted in the case of German federalism, the Länder are more or less responsible for the delivery of policies and programs and the Federal government is responsible for setting out the policy and regulatory frameworks.
Institutional design here is critical. You might well ask: what stops the Federal Government in Germany from just taking over services delivery? The public hospital system, for instance? And what stops the Federal Government from just imposing unreasonable and unaffordable policies on the Länder? The short answer is that the German Constitution does. The Constitution says that only the Länder can carry out administration in areas where they have jurisdiction. The Constitution also sets up a system where the Länder have to agree to spending requirements before they can be passed by the National Parliament. The Bundesrat, the "federal house" of the German Parliament, is composed of the governments of all the Länder. It is as if Mr Brumby, MrIemma etc. all sat in the Senate in Canberra and had the right to vote on any laws or policies that affected Victoria or New South Wales.
The Bundesrat system means that the Länder have a real veto power over policies and programs that affect their priorities and spending. This is what I mean by institutional design being integral to cooperative federalism.
My third observation has to do with the most significant weakness of cooperative federalism. The German system of federalism and the power it gave the Länder was in part a reaction to the Nazi era and the way the Nazi's emasculated the federal system and centralized power. It is also a system that suited the consensus and even "corporatist" style of post-war government in Germany. Where all sectors of society and all sides of politics had the clear imperative of reconstruction and rebuilding the economy.
With reunification and the increasing globalisation of the economy in 1980' s and 1990' s; not to mention the emergence of Brussels as a dominant source of authority, especially in Germany; the old consensus began to breakdown. There were increasing differences in interest between left and right, labour and capital, between regions and east and west Länder. There were also important issues about reforming the pension system, the health system, the schooling and university systems. And industrial relations.
It became clearer and clearer that the institutions of cooperative federalism were not able to deliver these reforms. The Germans used the term "Verflechtung" to describe the situation. They meant that the interests of the Federal Government and the Länder had become so "entangled" and "enmeshed" with one another that it was impossible to get anything decided or determined. It was especially impossible for any level of government to strike out on its own and try something novel or innovative. Everyone had to be in it together.
The Germans are currently engaged in a series of difficult attempts to reform their system. To "disentangle" the roles and responsibilities of the Federal Government and the Länder. So that each has greater autonomy over certain areas of activity. ("Entflechtung") They are, in essence, attempting to shift the German system more in the direction of a "coordinate" system of federalism like Canada or the US.
My fourth observation has to do with money or revenue. I would hazard the hypothesis that a coordinate system of federalism requires a reasonable level of fiscal autonomy. In other words, if you want a system of federalism where States have a fair degree of policy autonomy, then you are going to have to make sure that they also have a fair degree of fiscal autonomy. That is certainly true of US, Canada and Switzerland. Conversely, and perhaps more importantly, if you have a system where States do not have fiscal autonomy, then you will inevitably need to move in the direction of cooperative federalism. That, I think, is what has and is happening in Australia.
It is a notorious feature of Australian federalism that the States have virtually no revenue base to speak of. The States raise only some 50% of the funds they spend. The rest comes in transfers from the Federal Government. And some 50% of that comes in the form of tied grants. Clearly, tied grants are a way of controlling what the money is spent on. The Federal Government says "You can have this money on condition you spend it on these things or in this way." What is rarely commented on is the fact that the Federal Government, as a matter of good governance, is obliged to account for the way its tax dollars are spent. Commentators mainly concentrate on the use of funds as a means of controlling State policies. But there is equally an argument that a Federal government that transferred funds without taking an interest in how the funds are spent would be acting irresponsibly.
Ok. So what about the 25% of funds that are transferred unconditionally? This is the money that comes from the GST. Under the agreement entered into between the States and the Howard Government, the States agreed to retire some of their inefficient taxes and were given exclusive access to the GST revenues. Costello was always at pains to claim that the GST was really a "State" tax. It was just that under the Constitution (Section 90) only the Federal government can levy an "excise", which is what the GST is. The Costello claim, however, is a fiction. Over time, there will be more and more pressure on Federal governments to influence or determine how GST dollars are spent. Or, the other side of this coin, to account for how Federal tax dollars are being spent.
If you consider the most recent agreement between the Federal Government and the States about Special Purpose Payments or "tied grants", you begin to see how what I am predicting is already being played out.
What the Premiers and the Prime Minister agreed was to establish a more outcome focused system of fiscal transfers. Instead of focusing on the way in which money is used, they would focus on achieving certain results. They would agree on key performance indicators for certain areas of policy. Presumably things like waiting times for medical procedures, levels of morbidity, readmission rates etc. Outputs or outcomes.
They also agreed that rather than having a proliferation of funding agreements, they would amalgamate and broaden those funding agreements. So the agreements would be fewer and wider in scope. This has the potential to get rid of red tape and to get rid of perverse incentives and artificial constraints on policy tools set up by narrow boundaries.
So far so good. But if this system is to function properly, the KPI's or key results in say the health and aged care area, are going to have to be comprehensive. It is no good having a health system that concentrates just on waiting times, or just on Aboriginal morbidity, or just on diabetes. That would almost certainly create a skewed set of incentives, perverse incentives. So, the States are really going to have to agree with the Federal Government about the whole of health policy, all aspects of the systems and report against a comprehensive set of KPI's. Even though the Federal Government may be providing only 50% of the funds for health, they will have influence over the entire area of health policy. All areas of health.
You can see how this also entails that the Federal Government will have influence over all the funds spent by the States on health. Not only the "tied funds" but also the "untied funds". And that is not a perverse outcome or some sort of "trick" by the Federal Government. No, I would say it is an inevitable and inexorable consequence of the lack of fiscal autonomy by the States.
Let me say a few words about what happens in Germany. On the face of it, the German Länder have much greater fiscal autonomy. Taxes are generally shared between the Federal Government and the Länder. The share is determined by a formula that is set out in law. And most commentators on Federalism would say that the degree of vertical imbalance is nowhere near as problematic as it is in Australia. But in a sense, it is actually worse. Why? Because tax law reform requires consensus. In Germany, no jurisdiction or all jurisdictions have responsibility for Germany's tax system.
What the Germans are currently discussing is the very difficult issue of disentangling the tax system. So that the Länder will have more fiscal autonomy and responsibility for revenue and so will the Federal Government. That is very much work in progress.
The final observation I want to make about cooperative federalism ties in with my observations about the recent changes agreed between the Prime Minister and the Premier. And that is this: cooperative federalism, any sort of federalism, only makes sense to the extent that the States have freedom to do things differently, to experiment, innovate, even make mistakes independently of one another. To the extent that they all act uniformly, all do the same thing, there is no point in federalism. You might as well have a unitary system.
We had this discussion recently in Brazil where I was discussing their "tax wars". Brazil's system of VAT tax is a mess. There are essentially different and competing tax systems in each State. But worst is the fact that the States are competing their taxes away to attract industry and development.
They looked at different reforms to the VAT tax system. One option was to do what Australia did: have a single national VAT tax imposed by the Federal Government. Another was to have each State put in place the same VAT tax law. The end result in both cases was the same. The second option was "higher maintenance". And we were left wondering what the point of a "federal approach" to taxation was in this case. In the end, the preferred option that is being put before the Congress does give more freedom to the States to compete on tax rates but to have a uniform base and set of definitions. It does, in this way, preserve some point in having a federal approach to VAT.
In Australia's case, the Federal and State Governments are going to have to be careful to give the States "space" and incentive to innovate and compete. They are going to have to give the States room and leeway to adapt policies and programs to suit local circumstances and conditions. To strait jacket the States and Territories is to remove the whole point of federalism. It is, in a sense, to do away with federalism.
So, the KPI's are really going to have to be framed to concentrate on outcomes. They are really going to have to be framed to maximise room to manoeuvre, to try to achieve these results in different ways. On the other hand, the KPI's are also going to have to be meaningful. Not vague quantitative measures. But hard quantitative standards if they are going to do the work of driving reforms.
So, let me sum up the key lessons for cooperative federalism -:
- Cooperative federalism still requires sharp boundaries. It is not a matter of everyone being responsible for everything. The boundaries are, however, about the roles and responsibilities within an area of policy, not between areas of policy.
- Cooperative federalism requires good institutional design - you need to avoid the consensus trap but ensure the States are not handed "unfunded mandates".
- Cooperative federalism still has to take subsidiarity seriously - ensuring that States are given as much freedom as possible to do things differently and to respond to local differences.
- Addressing the issue of vertical fiscal imbalance would improve the functioning of the federal system, by making the States more accountable for the revenue they raise. But it might also remove the chief engine of cooperative federalism in Australia.
Roger Wilkins AO
11 September 2008