What follows is former Finance department head Michael Keating‘s speech today to the Australia Institute’s Revenue Summit 2023, calling on the Albanese government to substantially increase taxes to keep pace with public expectations for service delivery.
The starting point for any discussion of public policy should be the purpose of that policy. Only after agreeing on that purpose can we then begin to analyse the effectiveness of the policy in achieving its purpose.
That seems pretty obvious, doesn’t it? But in the case of taxation that is not the way most discussion proceeds͘ Instead, the starting point is typically that taxation is a burden, with the underlying assumption that this burden must be minimised͘.
Thus, over the last decade, the distinguishing feature of the governing Coalition’s tax policy was that it set an upper limit for its revenue equivalent to 23͘9% of GDP͘ No real attempt was ever made to justify this upper limit — it was just an arbitrary assertion!
And the new Labor government is not much different͘ While Labor has no explicit revenue target, and while it is generally sympathetic to the various demands for better services, Labor insists that its responses must be limited by what can be financed by much the same taxes as in the past͘.
However, in this talk, I want to argue for a fundamentally different approach to taxation͘.
Instead of setting an arbitrary limit to taxation, my starting point is to consider the purpose of taxation͘. That purpose is, of course, to pay for a range of services that governments either provide themselves directly or which they are expected to at least partially finance͘.
What services should governments fund?
But even if we agree in principle that the purpose of taxation is to fund government services and assistance, we then need to agree on what services governments should be responsible for and who should pay.
Reaching such an agreement has not always proved easy in the past. Historically, the role of the state and its service responsibilities have evolved, with major differences between different political parties and philosophies͘.
Originally, the only people who had the power to determine the role of the state were property owners͘. Consequently, the state’s role was essentially limited to the protection of property rights. This, in turn, meant that taxation was only required to pay for the defence of the realm and the justice system.
But with the spread of democracy and universal voting, the demands on the state inevitably increased.
For a long time after the introduction of universal suffrage, the responsibilities of the state were a key contested issue in most advanced democracies. Conservatives could promise to reduce taxes by refusing to fund some services or limiting entitlements to assistance.
But post the Second World War, there was general agreement to expand the role of the state in terms of income support and health and education.
The UN Declaration of Human Rights promulgated immediately after that War asserted that “Everyone has the right to a standard of living adequate for the health and well-being of himself and his family, including food, clothing, housing and medical care and necessary social services, and the right to security in the event of unemployment, sickness, disability, widowhood, old age or other lack of livelihood in circumstances beyond his control”.
This UN Declaration both reflected and encouraged a new consensus about the expansion of the role of the modern state. Nevertheless, the speed and extent of this transformation in the role of the state were debated for the next thirty or more years.
Today, however, I think there is a large amount of consensus concerning the role of the state and its expenditure responsibilities. We recognise that all Australians are entitled to basic levels of education, health care, income support and shelter and that governments have a responsibility to ensure the provision of these essential services.
Furthermore, there is now widespread acknowledgement that these services that are financed collectively are critical to building our community and our sense of community. Indeed, as the great American jurist, Oliver Wendell Holmes, said: “I like to pay taxes. In this way I buy civilisation.”
Thus, taxation reflects our mutual obligation to one another as citizens. Taxation underpins an inclusive society and is an efficient way of paying for those services that are collectively consumed. Moreover, many of these services paid for by taxation add to our quality of life and history suggests that our demand for these services, such as health and education, tends to rise relatively fast as our incomes increase through economic growth.
The problem for governments, however, is the present disconnect in the public mind between the people’s demand for services and the taxation to pay for them. In effect, we are schizophrenic. We want increased access to more and better services on the one hand and less taxation on the other.
But the response to this dilemma cannot lie in cutting expenditures by reducing government responsibilities.
Most obviously the last such attempt by an Australian government to roll back some responsibilities was the new Coalition government’s first Budget in 2014. However, the popular reaction against these savings measures was so strong, that the government quickly withdrew them.
Following that defeat, Coalition governments were not prepared to incur the wrath of the electorate by seriously reducing government responsibilities.
The Coalition learnt its lesson and did not challenge the public consensus on the range of services and assistance that the government was expected to provide. Instead, the Coalition relied on stealth by underfunding many programs to achieve its Budget targets͘
But even with that underfunding, the Coalition could not achieve its budget targets. In its last Budget before the impact of COVID-19 total payments were 24͘5% of GDP, significantly more than the Coalition’s revenue ceiling.
Is government service provision presently inadequate?
However, even if we are largely in agreement on the role of the state and its service responsibilities, the Albanese government does not appear to consider that present tax revenues are inadequate to finance those responsibilities.
The recently released Intergenerational Report (IGR) projects that, if the government’s existing policies are maintained, payments will increase over the next ten years from 24͘8% of GDP to 26͘6% in 2032-33. Total receipts are projected to increase from 25͘0% of GDP in 2022-23 to 26͘3% in 2033-34. The net impact is that the Budget deficit will be below 1% of GDP for most of the next ten years and fall before it starts to grow relative to GDP from the mid-2030s onwards.
But this projected Budget deficit may well be optimistic. The IGR projections assume that productivity will increase at an average rate of 1͘2% — much faster than over the last ten years from 2012 to 2022 when the average annual rate of productivity growth was only .0͘5%. Furthermore, this slowdown in productivity was common to most developed economies, and the extent to which it can be reversed is questionable, to say the least.
But if productivity growth is slower than projected, then budget revenue will be less, and the deficit will be larger than projected.
Nevertheless, the tone of the IGR and associated government statements imply that the government is not concerned by its projected deficit and the risk that it could be even higher. The clear inference is that the Albanese government considers that present taxation arrangements are adequate to cover the cost of its existing policies.
But the problem is that those existing expenditure policies are effectively governed by the existing taxation arrangements. Thus, service provision is being limited to what can be accommodated by those existing taxation arrangements. Unfortunately, there are good reasons to doubt that this level of service provision is adequate, including most notably in the following areas of service provision.
First, while the government has responded to the Royal Commission into Aged Care and expanded access to childcare, it is arguable that both remain underfunded. In addition, the increase in the availability and quality of care services is likely to lead to a rise in their cost because of the wage increases required to attract the necessary numbers of staff.
Second, the government has also increased the funding for social housing, but again this extra funding is widely considered to be much less than what is needed.
Third, Australian government hospital funding is projected to grow over the next four years at an average annual rate that is 1% faster than over the previous nine years under the Coalition͘ That faster increase is to be welcomed but it is questionable whether that will be sufficient to reduce the waiting times that are presently excessively long.
Similarly, the funding for primary health care services in Labor’s latest Budget includes additional funding for a range of measures to strengthen Medicare as recommended by that taskforce. But the Medicare rebate is still very low, risking a growing shortage of GPs, particularly in low-income and rural regions where the patients cannot afford large gap payments.
Fourth, over the nine years of the Coalition government real funding for tertiary education hardly increased and it fell per student. Over the next four years under Labor this funding is projected to increase at an average annual rate of 3͘8% in nominal terms.
That effectively means that Labor is planning not to restore the past cuts in real funding per student in higher education, and it may even fall a bit further. While in the case of VET, the Albanese government has already acknowledged that its existing policies were inadequate. Since the release of the IGR, the government has released a new National Skills agreement where the federal government will contribute an annual average of $2͘5 billion. This is significantly higher than the average of $1͘6 billion per year under the previous policy and will thus add to the projected Budget deficit.
Fifth, Australia’s level of income support for unemployed people is very low compared to most other developed economies. Similarly, increased rent assistance should be a priority, especially given the recent rise in rents.
Sixth, the deterioration in international security means that an increase in defence spending and foreign aid is likely to be needed. Defence experts argue that to accommodate the cost of the AUKUS nuclear submarines, without jeopardising other defence capabilities, the Defence budget will need to increase by at least 1% of GDP over the next decade or so.
In addition, foreign aid has been seriously underfunded for a long time. In 2022-23 Australia’s overseas development assistance was 0͘20% of gross national income, down from 0.33% under the Gillard government in 2011-12, and way below the OECD average of 0.32%. Realistically, our diplomatic effort to improve our security also requires a dramatic lift in foreign aid.
Seventh, action to reduce carbon emissions and to fully develop Australia’s potential as an alternative source of energy will almost certainly require greater funding than is presently being provided under existing policies.
After making a modest allowance for additional spending in response to these demands, the Grattan Institute thinks the structural Budget deficit will be close to 3% of GDP in ten years if present tax arrangements continue to prevail.
My own personal assessment is that, if the government meets public expectations for adequate services and income support, then unless there is offset action to reduce some other expenditures or to increase taxation, the budget deficit would be closer to 4% of GDP.
Furthermore, I think that trying to finance government service provision by increasing budget deficits and borrowing is quite unrealistic.
The Budget balance should always be determined by the need to offset the difference between private savings and investment͘ In Australia, it is likely that this difference will require a balanced budget or even a small surplus whenever there is full employment, which is of course the aim.
The possibilities for expenditure savings
However, before arguing for higher taxes, the other alternative would be to reduce some expenditures. Major savings would, however, require policy changes that will reduce services, but I doubt the willingness of governments to make such policy changes.
Less controversial are savings from improving program efficiency, and incoming governments often promise that. While any such opportunities should be investigated, I doubt that much can be expected from any efficiency reviews.
In its first Budget, the Albanese government did find expenditure savings equivalent to ½ per cent of GDP. But that is not much, and presumably, these were the easiest savings, and even then included some modest policy changes.
That reinforces the doubts that much more can be done to reduce Budget outlays without significant policy changes to reduce some services or assistance.
In fact, I estimate that two-thirds of Australian government expenditures relate to entitlement programs, where the entitlement is specified in the legislation, or represents high priorities such as defence or public hospitals͘ The administrative costs of the entitlement programs are negligible so efficiency gains cannot be expected there͘. While, after a decade of underfunding by Coalition governments, I doubt that much in the way of savings from increased efficiency can be achieved in the other high-priority programs.
That still leaves a third of Australian government expenditure where potential savings might be easier. However, the aggregate spending on these other programs was already projected in the October 2022 Budget to fall at an average annual rate of 7͘8%, which suggests that further savings in these programs would be very difficult too.
The one obvious exception is government investment in new infrastructure, where the vast majority of projects have been agreed upon by governments without any evaluation. But the reason is probably because the politics of infrastructure investment override what governments expect would be a negative evaluation.
In any event, while a more disciplined approach to infrastructure investment should be welcomed, this is still unlikely to result in major Budget savings. The total expenditure on road and rail transport represents only 2% of the Australian government’s total budget outlays, so even halving that investment would not be anywhere near sufficient to make up the shortfall in other program expenditures.
No doubt conservatives will disagree and continue to assert that they can lower taxes financed by reducing government outlays. They should be pushed to say where, but they won’t because they can’t. Such claims that taxes can be lower should therefore not be taken seriously.
Taxes and economic growth
The other argument that conservatives will employ in favour of lower taxes is that it will accelerate economic growth. These supporters of low taxation typically argue that tax cuts will incentivise people and businesses. Indeed, it is not unknown for some politicians to assert that tax cuts will increase our economic growth sufficiently to cover the cost of the services that we demand.
Of course, we can all think of specific taxes, that if they are levied at a high rate, will change behaviour; indeed, that is the intention of taxes on cigarettes and carbon, and charges on congestion. But the empirical evidence shows that there is no correlation between present overall levels of taxation and any country’s economic growth rate.
Many high-taxing countries in northern Europe for example, have a higher growth rate in their per capita GDP than low-taxing countries such as America. Similarly, the drop in Australian income tax rates in the 1980s was not associated with any increase in employment participation or productivity growth. And Australia is already one of the lowest taxing countries among the developed nations of the OECD.
Instead, what really matters is what a country does with its taxation revenue. If it is spent wisely on functions, such as research and development, education and training, health, and infrastructure, then this taxation can actually increase the nation’s economic capacity. For example, that is why, even within the same country, such as the US, the economic performance of Massachusetts has been better than that of Mississippi, although Mississippi has lower taxes.
Even if Australia increased its taxation revenue by as much as another 4% of GDP, as I have suggested is needed, that would not harm economic growth. Taxation in Australia would still only be raising about the same amount of revenue relative to GDP as the OECD average, and Australian taxation would still be lower than in Canada and the UK, and well below the average in the 17 Euro countries.
How to change the political debate
In short, the case for increased taxation right now in Australia depends upon establishing that existing government policies will not adequately fund the true cost of providing adequate services and assistance to the needy.
But how do we get this message across, and hopefully change the political debate in favour of popular support for increased taxation?
Fundamentally, we need to improve the link in the public mind between its demand for services and the taxes necessary to pay for them.
Ideally, the government would admit the challenges in adequately financing the provision of essential services. The aim would be to set the scene for a full-scale independent public review of the funding required for adequate service provision. This review could also build upon the findings of other expert reviews such as the various Royal Commissions, the Medicare Task Force, the Economic Inclusion Advisory Committee, the Women’s Economic Equality Committee, and the Defence Force Structure Review and the follow-up defence reviews that have since been commissioned.
The aim would be to achieve an informed assessment and public debate about the range and scale of the services to be provided and their cost. Once that amount of desired public expenditure has been authoritatively established, the government would then be on much stronger ground to argue that that amount of expenditure should determine the amount of tax revenue that therefore needs to be raised.
Sources of extra revenue
Hopefully, this authoritative estimate of how much revenue is really needed would then shift the national conversation away from demands for lower taxes to a discussion about how best to raise that required revenue.
I don’t have a lot to contribute to this discussion about new revenue sources. To me the critical issue is to get agreement on the need for more revenue; how that additional revenue is obtained is a secondary consideration.
Also, others are more expert about how best to change the tax mix͘ Indeed, probably the best starting point for this discussion is still the expert Henry Report written a decade ago.
Nevertheless, I will offer a few thoughts͘
First, a high priority would be to rely more on those taxes which would improve efficiency, such as a carbon tax, congestion charging, and substituting a land tax for stamp duty.
Second, Australia raises relatively little from its GST compared to other countries. It would be possible and arguably desirable to raise more by a combination of broadening the GST tax base and raising the GST tax rate.
Third, a resource rent tax so that the community shares some of the super profits that mining companies can make due to fortuitous circumstances rather than their effort or initiative.
Fourth, the avoidance of company tax has become an industry. As I said, I am not an expert, but I would urge close consideration of Ross Garnaut’s proposal to substitute a tax on corporate cash flows for the present company tax.
Fifth, loopholes in the personal income tax system, such as the inadequate rate of capital gains taxation and negative gearing should similarly be closed.
Sixth, and finally, we need a more progressive income tax rate scale. The starting point should be to at least adjust the Stage 3 tax cuts, but a full-scale review of the rate scales would be better.
As things presently stand, if we combine the three stages of the previous government’s tax reform package, modelling by the Parliamentary Budget Office shows that in 2024-25 a huge 64% of the total cost of the package goes to the top income quintile of people lodging an income tax return, 20% to the fourth quintile, but less than 13% to the middle quintile and almost nothing to the bottom two quintiles.
Furthermore, a projection of this tax package shows that, because of bracket creep, by 2030-31 the lower four quintiles of taxpayers by income will have a higher average tax rate than they had in 2017-18͘. Those in the second and third quintiles (low to middle-income earners) will face the largest increases of about 4 percentage points in their average tax rate. By contrast, the average tax rate for those people in the top income quintile will be slightly lower.
This shift to a less progressive income tax system is inequitable and will be counter-productive in a society where inequality has already increased and is the prime cause of economic stagnation.
On the other hand, it is not hard to think of good ways to raise the additional tax revenue that we will need to sustain a decent civilised society.
Now to sum up.
There is ample evidence that government services and assistance in Australia are presently underfunded, and service provision is therefore inadequate.
The only way to fix this is to increase the amount of taxation revenue. But this will require an important change in government policies, which are founded on almost no significant change to taxes.
And while the government seems to be sympathetic to many of the demands for improved services, it is most unlikely to change its tax policies without popular support.
So we need to change the public debate. The public needs to overcome its present schizophrenia and accept that you cannot have adequate service provision without paying more taxes.
Accordingly, my proposal is an authoritative review and assessment of how much revenue is needed to adequately fund all government services. When the public has a better understanding of what they will get for their money, and how much is needed, then hopefully they will be more prepared to accept the need for increased taxation revenue.