When will the Leviathan fade away?
Des Moore
Click here for PDF version
The federal government has failed to curb discretionary spending, argues Des Moore
The very best of all plans of finance is to spend little, and the best of all taxes is that which is the least amount.
Jean-Baptiste Say
Nobel Prize winner James Buchanan argued in the 1990 John Bonython lecture that, while it took about a century for the widespread faith in socialism to die (from mid nineteenth century to between the early 1960s and the early 1990s), there remained a residual unwillingness ‘to leave things alone’. He perceived that natural forces would generate the Leviathan state and the ‘self-same interests that Adam Smith sought to abolish are everywhere resurging, as if from the depths of history’.
Sixteen years on Buchanan would doubtless confirm his assessment were he to return to Australia today. Even though avowed believers in ‘small’ government have been in Canberra for more than ten years, total discretionary government outlays in Australia have increased by two and a half percentage points of GDP since his lecture (in this context ‘discretionary’ is all outlays less interest payments). Politicians are today spending the equivalent of $23 billion more than their predecessors would have with the size of government when Buchanan delivered his lecture.
Such a spending pattern lends support to Buchanan’s conclusion that ‘if we know that politics fails and that its natural proclivity is to extend its reach beyond tolerable bounds, we may be led to incorporate constraints into a constitutional structure’. But is it likely that politicians would propose rules, particularly constitutional ones, to constrain their own natural inclinations to help attract votes?
Developments overseas
Buchanan might be less concerned by developments in the size of government in Europe. Surprising as it may be to some, the proportion to GDP of general government outlays by all forms of government at least show noincrease since 1990 for either Europe or the OECD as a whole, and small decreases in the case of each group of countries since 1995.[1] However, as with Australia, these post-1995 decreases are probably due mainly to reduced interest costs rather than lower discretionary spending.[2]
Nevertheless, there are some mildly encouraging signs of movement in the direction of smaller government. Since 1995 the weighted average outlays of all OECD countries has fallen from 42.1% to 40.7% of GDP (still nearly 6 percentage points of GDP higher than for Australia) and no less than 21 out of the 28 countries surveyed have reduced their outlays (the reduction in the OECD average would have been greater but for the relative stability of outlays in the bigger countries such as the US, France and the UK).
Importantly, twelve of these countries have reduced outlays by five or more percentage points of GDP, which suggests that they have made reductions in discretionary outlays. Particularly noteworthy have been the reductions since 1995 by the Nordic ‘big spenders’—Sweden (11 percentage points), Norway (9), Finland (8) and Denmark (7)—as well as those by Canada (9), Czech Republic (10) and the Slovak Republic (9).
Of course, the downward trend in these smaller OECD countries still leaves them with significantly larger government outlays than in Australia, but there is now a smaller gap. For example, Canada’s outlays of 39% of GDP are now less than five percentage points above Australia’s compared with more than ten in 1995 and 14 in 1990. And notwithstanding the temptation to splurge its oil revenue, Norway at 42% of GDP has reduced its gap to eight percentage points compared with 19 in 1990 and 13 in 1995. Hopefully, the reductions in the Nordic countries indicate that countries that developed big, socialist governments have started tentative steps away from big government philosophies.
Recent developments in Canberra
Regrettably, there is no sign of any similar movement towards smaller government in Canberra. This is despite denials by senior ministers of any increase in the federal components of outlays and taxes and claims that since 1996 the Coalition has effected a major reduction in the size of government.
In an address to the National Press Club on 24 August 2005 Finance Minister Nick Minchin asserted that, over the life of the Coalition government, ‘total federal government spending had fallen as a proportion of the total economy from 25.5% of GDP to just 22.1% in 2005–06’. Treasurer Peter Costello wrote in The Age on 27 March 2006 that ‘from 1995-96 through to 2005–06 the Australian[3] Government’s spending has declined from 26.3% of GDP to 21.6%’ and claimed that ‘one of the values that Liberals hold dear is to be disciplined with government spending’.
And, in a speech entitled ‘Taxation: Keeping Faith with Australian Families’, Prime Minister John Howard told the Menzies Research Centre on 18 April 2006 that ‘the Australian Government’s overall tax share has fallen as a proportion of GDP from 23.1% in 1996-97 to an estimated 21.0% in 2005-06’, adding that ‘we have the eighth lowest tax to GDP ratio in the OECD’. But, while the OECD Factbook for 2006 does show the total Australian tax burden is the eighth lowest, he omitted to mention that it also records an increase of nearly two percentage points between 1995[4] and 2003 (and it will have increased further since then).
There is little doubt that if a CEO of a major company had falsely claimed his company’s costs had fallen over recent years by a similar proportion to the ministers’ claimed reductions in spending and taxes, he would be subject to heavy criticism and hauled before the regulatory authorities. Yet neither the Opposition nor media commentators have actively pursued the issue, although The Age (Business Section) did publish my letter to Minister Minchin, which pointed out his incorrect Press Club comparisons, as well as his evasive reply. Nor have I been successful in my attempt to persuade the Treasury to publish an appendix in Budget Paper No 1 showing the correct figures already available on the Australian Bureau of Statistics web site.
The main problem with the ministers’ figures is that the pre 2000–01 data include large general revenue grants to the states and the taxes to pay for them while the ones for recent years exclude from expenditure the GST payments now made in lieu of those grants and exclude from taxes the revenue from the GST itself. This means that none of the ministers’ comparisons has a valid, let alone a meaningful, basis. However, because the ABS (bravely) decided to adopt the correct approach of treating the GST as a federal tax and its payment to the states as federal expenditure, its web site provides data on Commonwealth spending and taxation on a consistent historical basis.[5]
So much, then, for section one of the Charter of Budget Honesty Act, which states that its purpose includes ‘facilitating public scrutiny of fiscal policy and performance’. The public has been seriously misled by these ministers, who should make a full explanation and public apology for presenting such a distorted picture of federal fiscal performance. The Charter is badly in need of amendment to ensure accurate data is published.
My May 2005 report to the Australian Chamber of Commerce and Industry on Commonwealth Spending (And Taxes) Can Be Cut—And Should Be had made the situation clear even before Finance Minister Minchin’s address. I have now updated relevant figures for both expenditures and taxation, again using published data from the ABS web site, which confirm my analysis. Some important years for comparative purposes are as follows:
This shows that, while there has been a decline in total federal spending of about one percentage point of GDP since 1995–96 (ie, much smaller than claimed by ministers), this decline is all due to lower interest costs. Moreover, although spending did reach a (Coalition) ‘low’ in 1998–99, there has since been no sign of moving towards smaller government. Indeed, discretionary spending by the Coalition has actually increased by nearly one percentage point of GDP since 1995–96.
The Budget Measures document for 2006–7 reveals the hand-out proclivities of the Coalition. Decisions made since the 2005–06 budget are estimated to cost a total (including tax cuts) of some $47 billion net over the three years to 2008–09. The document describing the decisions taken is the largest ever, comprising 357 pages and explaining close to 1,000 new expenditure measures alone. This kind of behaviour smacks of government losing control.
True, the $47 billion includes $28 billion of the $37 billion in tax cuts (spread over four years) so the addition to net spending is ‘only’ about $20 billion. But even these tax cuts are not ‘real’ in any lasting sense. Retention of the progressive tax scale means that revenue soon recovers the loss from income tax reductions and by 2009–10 bracket creep will have restored as big a proportion of GDP in income tax as the government received in 2005-06. Taxpayers have thus been given a small, short breather but there is no permanent lowering of the income tax take relative to GDP.[6]
Indeed, even with the tax reductions it is clear that the Coalition is the heaviest taxing government we have ever had in Canberra. In 2004–05 the tax burden imposed by the federal government was about three percentage points higher than in 1995–96. Even on the improbable assumption that all the $10 billion ‘black hole’ left by Opposition Leader Kim Beazley in that year would have had to be filled by higher taxes, the resultant tax burden would still have been about one percentage point of GDP higher. Under the Whitlam government the total tax take was 8 percentage points of GDP lower, equivalent in today’s terms to a tax cut of over $70 billion. In short, the Australian experience under the Coalition seems to confirm Buchanan’s Leviathan thesis.
In assessing spending proclivities, account needs to be taken of the Tax Expenditures Statement (TES) which lists the very large number of concessions, benefits or incentives provided through the tax system and identifies the net revenue cost by comparing the revenue that would be collected if the standard tax treatment was applied. The TES for 2005 (the latest) runs to over 160 pages (again the largest ever) and reveals an estimated total cost of 4.1% of GDP (or $37 billion) in 2004–05 and each of the years to 2008–09.[7] As acknowledged in the TES for 2004, ‘tax expenditures are substitutes for direct expenditures, delivered through the taxation system and accordingly affect the budget position’. Accepting this Treasury acknowledgement, it could be said that Commonwealth outlays for 2004–05 are equivalent to nearly 30% of GDP, that is, to the 25.2% derivable from ABS data plus 4.1% of tax expenditures. Correspondingly, it could also be said that the total Commonwealth tax take is nearly 30% of GDP.
Despite the large sums involved, and their obvious economic/social policy implications, these tax expenditures do not receive regular scrutiny through the budget process. Yet they include large amounts in 2004–05 of concessional taxation for funded superannuation ($14 billion), the exemption of family tax benefits ($2.3 billion), the discount of capital gains for individuals and trusts ($4.3 billion), the tax offsets/exemptions of various income support benefits, pensions or allowances, the tax offset for senior Australians and the application of a statutory formula to value car benefits.
Account also needs to be taken of Australia’s superannuation arrangements, under which employers are legally liable to pay a superannuation guarantee charge equivalent to a 9% payroll tax if they do not make a superannuation contribution for their employees. Although ‘forced’, this contribution is not treated as a tax—because it is not paid into government accounts but straight to superannuation funds—and the resultant payout from super funds to contributors is thus not included in government expenditure. By contrast, in European and North American countries a higher proportion of pensions are financed by payroll taxes paid to governments by employers and employees and the government payouts to contributors are thus counted as public sector expenditure.[8] As the contributions by Australian employers (equivalent to about 3.5% of GDP) become, arguably, a de facto addition to Commonwealth outlays, the size of government in Australia is thus much closer to the OECD average than appears to be the case.[9]
Australia’s private sector role sets a good example
Although some commentators criticise our ‘small’ government sector, this fails to recognise the higher proportion of services provided through the private sector in Australia. Total Australian spending (that is, both government and private) on services such as education and health is broadly in line with total spending in other countries with comparable income levels. It should be acknowledged that our larger private sector is an advantage in that it reduces the adverse effects of taxation on productive effort and, importantly, also means that a more efficient and higher quality service tends to be provided.[10] Far from ‘worrying’ about the small size of our government sector, we should be flaunting its benefits and looking for ways to reduce it further.
It is relevant that the dividing line between the private and government sectors has ceased to be clearly defined because of the increasing realisation in the community that the private sector has the capacity to provide a large proportion of what have hitherto been regarded as public goods. In effect the natural monopoly rationale for the government to be a major supplier of goods and services is outdated. Although little recognised, Australia is already a world leader in usage of private sector education and health services, in the substitution of private for government services in areas such as banking, airlines, (some) electricity and (soon) telecommunications, and in the use of public-private partnerships.
For example, since 1997–98 the proportion of students attending fee-charging non-government schools has increased from 30% to 33% (with over one million students now attending such schools) and there has been a similarly large increase from 33% to 39% in the proportion of patients treated at fee-charging private hospitals. Apart from providing extensive savings to taxpayers, such developments reflect the higher quality and the wider choice that modern society wants.
Even though they have to pay fees as well as taxes, these users of private services that compete with government services have increasingly been voting with their feet. This should be sending a message to political parties.
Looking ahead
As to post 2004–05 developments, details of the budget outcome for 2005–06 will only be published by the ABS on its consistent historical basis after some delay. But, as table 3 shows, the 2006–07 Budget estimates hold out no promise of any change in direction. Indeed, while the figures for the out years (that is, those after 2006–07) are only projections rather than estimates, they offer a rather pessimistic scenario as they reflect existing policies which, on past experience, are bound to lead to spending creep.
An examination of the main categories of recent and projected trends in outlays suggests that the Prime Minister was wrong in his claim at the Menzies Centre that Australia’s government is ‘lean but not mean’—but ‘definitely not a nanny state’. Although no precise definition exists as to what constitutes a mothering-fussing government, the Coalition surely fits the bill with its supply since 1995–96 of assistance for welfare, health, and education at similar annual proportions of GDP, as table 4 shows, to a broadly unchanging component of the population. Of course, some of these outlays are justifiable in terms of social policy. But have they needed to increase more rapidly than average incomes, particularly in circumstances in which unemployment has fallen from about 8% to 5%? That this is pretty much what happened between 1995–96 and 2004–05, and is projected to recur again out to 2009–10, confirms that Canberra is very much into holding our hands.
While the Coalition government has now plucked up the courage to effect some tightening in eligibility for benefits such as disability support pensions, parenting payments, Newstart allowances, and pharmaceutical benefits, this is only reducing the growth in outlays at the margin and, in some cases, is actually increasing it. For example, although unemployment is projected to remain stable at around 5% over the period 2004–05 to 2009–10, assistance to the unemployed is projected to increase by 64%, presumably because of additional spending on training under the Welfare to Work program.[11]
Given his rejection of the notion of Australia as a nanny state, it is ironic that the Prime Minister’s favourite spending program is the family tax benefits scheme, now requiring about $28 billion a year (more than double what it was in 1995–96) and provided to over 3 million people, including all dependent children up to 21 years.[12] His justification? These benefits are, he told the Menzies Centre, not middle class welfare but ‘tax relief for a universal reality—that it costs money to raise children.’ If that isn’t ‘mothering’, it certainly looks very much like it!
Of course, helping lower income groups meet the cost of raising children is a legitimate function of government. But the universal reality of such costs for families does not provide justification for government assistance to all and sundry. A Coalition that should be expounding self-reliance should surely limit the eligibility of higher income households and set about progressively reducing the cost to taxpayers of family tax and many other benefits.
Support for such action is provided by the significant part of the tax and social security system that consists simp ly of churning taxes back whence they came, that is, to higher income groups themselves.[13] The latest official published analysis by the ABS of the distributional effects of government benefits and taxes relates only to 1998–99[14] and provides only an (ABS acknowledged) incomplete coverage of them. However, it does give a general indication of the extent of such churning by showing that households with incomes in the top two quintiles:
• received over 30% of all social security and ‘selected’ education and health benefits covered; and
• those benefits were equivalent to 47% of the taxes paid by them.
If the same proportions apply today, that would mean that about $45 billion of welfare, education and health benefits go to households with incomes in the top two quintiles. This surely indicates that considerable scope exists for reducing both taxes on and benefits for those households, with obvious economic benefits.
Where to now?
There is no proven direct correlation between changes in economic growth and the size of the public sector and some countries have produced poor economic performances even with governments that have relatively small outlays.[15] However, academic analyses do suggest a negative relationship can exist between the size of government and economic growth. In a recent book that includes a survey of a wide range of academic studies of the effects of government size, economics Professor Dennis Mueller of the University of Vienna concludes that, while too small a government sector can harm economic performance, beyond some point the adverse incentive effects from high levels of taxation and regulation outweigh its positive effects.[16] All of the highly developed countries, Mueller argues, are beyond the point.[17] But the country which has maintained one of the smallest government sectors since its inception—the USA—has real per capita income levels that are about 25% above the next highest in the OECD and is a country in which individual freedom is at a high level.
The case for reducing government outlays in Australia was given additional weight by the Intergenerational Report 2002–03, circulated by Treasurer Costello but clearly sounding warning bells from Treasury. The obvious implication is that the ageing of the population will in due course require a large increase in taxation if existing spending policies are continued. But, although this heightens the economic and social case for early action to reduce education, health and welfare programs, Treasurer Costello has done little to promote the necessary policies, viz the tightening of eligibility for obtaining benefits so that recipients are concentrated in the lower income groups.
From an economic perspective such policies would help meet the objectives of the Intergenerational Report by increasing the availability of labour and, with reduced labour market regulation, employment. They would also help lift the rate of personal saving from the very low level to which it has fallen. From a social perspective, welfare dependency would be reduced as an increased proportion of individuals and families would assume more responsibility for their own welfare. There would be less of a ‘nanny’ state.
But how can such action be achieved without the political leadership that would obviously be required but which is evidently absent at present? One contribution would be to at least promote debate on various relevant issues.
In the USA there has been an ongoing debate over many years on the idea of amending federal and state constitutions to limit the power of governments to borrow and tax. This doubtless prompted Buchanan’s reference in his Bonython lecture to possible constitutional constraints. Indeed many US states already have constitutional requirements to balance budgets. Term limits also exist for some elected representatives including of course the President.[18]
However, while term limits have some attractions, constitutional constraints on taxes and borrowing have inherent practical difficulties and any move in that direction in Australia would first require a change in the still wide culture of belief that a substantial government is needed because society cannot be left alone.[19] Although there are increasing indications that such a cultural change is happening—because experience has revealed that governments are not efficient suppliers of quality services—further substantive progress would require political leadership.
Accordingly, in addition to the need for private sector think-tanks to continue espousing the cause of smaller government, politically-inspired action to assist such a cultural change, and promote debate about the issues, might include:
• Initiating a major change in the existing presentation/structure of the budget papers so as to expand the information and analysis on the allocation of expenditure by function and classes of recipients, and on the sources of revenue and the classes of payees/concessionaries. The function of the budget has moved progressively over the past ten years or so from one that was primarily focused on the macro effects of changes in budget aggregates to one that is now (or should be) much more focused on the micro effects, viz the economic/welfare implications for particular groups and the role of government in society. There is now broad bipartisan agreement on desirable budget outcomes and that changes in aggregate expenditure and revenue have only limited effects on economic growth/inflation in the short term (using the budget to try to ‘fine tune’ the economy is also no longer regarded as an appropriate policy). Reduced attention could be given to the economic outlook, which could be the subject of a separate paper from Budget Paper No 1;
• An amendment to the Charter of Budget Honesty legislation to require the publication of spending and revenue data on a consistent accounting and historical basis;
• Requiring the Treasury or the Productivity Commission to publish papers analysing the changing role of the public and private sectors, particularly in regard to the provision of services hitherto regarded as primarily the responsibility of government, and an examination of the potential for increasing the competitive environment for service provision;[20]
• Requiring the Treasury or the Productivity Commission to publish an analysis of the reduction in outlays by most OECD countries in Europe.
Conclusion
Will it take a century to rid ourselves of James Buchanan’s Leviathan? There are some glimmers of hope suggesting it might be achievable in less time as an increasing proportion of the population recognises that government failure does occur, as they turn increasingly to private education and health services and demand that government-run corporations provide fewer services, and as public-private partnerships of one form or another are utilised. That Australia is in the forefront of such developments is also encouraging, as is the apparent shift towards smaller government in Europe, albeit from a higher level.
At the same time it is most discouraging that there has been no diminution in the last ten years in the overall level of government spending in Australia and that nanny type forms of assistance have continued apace. This is despite strong increases in real incomes and health. Political science has yet to find a solution to the temptations for office holders to buy votes.
What seems to be missing is a champion of the private sector and the individual. This is disappointing after ten years of a Coalition government that supposedly believes in smaller government and in giving greater responsibility to individuals, who are nowadays more able to take care of themselves. Let us hope that, in its policies for the next election, the Coalition will go one better than the Opposition’s pledge not to increase the tax to GDP ratio.
Endnotes:
[1]OECD Economic Outlook, May 2006.
[2] The OECD data, which appears to be compiled on a slightly different basis to the ABS data, shows a reduction between 1995 and 2005 in total Australian general government outlays of 3.3 percentage points of GDP compared with the 3.1 percentage point reduction shown above between 1995–96 and 2004–05. The OECD publishes no breakdown of its outlays data by function.
[3] In this context ‘Australian Government’ means the Commonwealth Government.
[4] As the Coalition took office in March 1996, 1995 or the financial year 1995–96 provide the best bases against which to measure the fiscal performance of the Coalition Government since being elected.
[5] The ABS data is also on an accruals basis whereas the ministers compare data from the pre-accrual and post accrual periods. The move from cash to accrual accounting occurred in 1999–00.
[6] The estimated $37 billion of tax reductions implies that total taxation over the four years will be about 4% less than otherwise.
[7] Budget Paper No 1 for 2006–07 includes a two page appendix on Tax Expenditures.
[8] In several European countries payroll taxes contribute a higher proportion to total taxes than income taxes.
[9] Of course, apart from taxes employers are forced to do more than pay superannuation contributions and, once government interventions in the economy are brought into consideration, the ‘size of government’ becomes a more amorphous concept. Even so, the superannuation contribution is a major government-imposed levy which is effectively unavoidable.
[10] In the CIS Tax Policy Monograph (8) on The Costs of Taxation, ANU Senior Lecturer in Economics, Alex Robson points out that ‘a dollar increase in government revenue ends up costing far more in real terms than the dollar paid in taxes … While tax revenues may be spent by government on beneficial uses, these benefits have to be weighed against the ‘deadweight losses’ (basically the value of lost output) incurred as a result of levying the tax in the first place’.
[11] On 1 July 2005, Treasurer Peter Costello issued a three page press release outlining details of tax and other benefits taking effect from that date. No mention was made of any eligibility tightening.
[12] The total cost is higher than this as the benefits are exempt from income tax at a cost of about $2.6 billion in 2006–07.
[13] CIS Social Research Director, Peter Saunders, has also written extensively about churning. See, for example, Pokie Principle a Winner, The Australian, 9 May 2006.
[14] ABS (Australian Bureau of Statistics), Government Benefits, Taxes and Household Income, 1998-99, ABS Cat. No.6537.0 ( Canberra: ABS, August 2001). The next edition will be for 2003–04 and is to be published in 2007.
[15] In Why Globalization Works (New Haven: Yale University Press, 2004), leading economic journalist for the Financial Times, Martin Wolf, points, on the one hand, to the recent poor economic performances of countries such as Japan and Switzerland with relatively low government outlays and, on the other hand, to the fairly strong economic performances of Sweden and Finland with relatively large government outlays. However, as noted, the latter two countries have been effecting major reductions in the relative size of outlays and Japan’s relatively poor performance can be attributed to the extensive government intervention in the economy apart from government spending.
[16] Dennis Mueller, Public Choice III (Cambridge: Cambridge University Press Cambridge, 2003).
[17] See also the April 1998 report by the Joint Economic Committee of the US Congress on ‘The Size and Functions of Government and Economic Growth.’
[18] There are various limits on terms of Governors in 36 states and in 1995 the Republican Party secured a simple majority in the House of Representatives limiting House members to six two year terms and members of the Senate to two six year terms. However, as a two-thirds majority is required for a constitutional amendment, the bill failed.
[19] In 1994 there was an attempt in Australia when the NSW Government announced that it would legislate to guarantee a balanced budget.
[20] For further analysis of this approach, see my report to Commerce Queensland on The Role of Government in Queensland, May 2006, www.ipe.net.au.
Des Moore is a former Deputy Secretary, Commonwealth Treasury and is now Director,
Institute for Private Enterprise.
|
