Printer Friendly

Dracula in charge of the blood bank.

1. Introduction and Summary

An Australian pension review was chaired by Dr Jeff Harmer when he was Secretary of the Department of Families, Housing, Community Services and Indigenous Affairs. The report was published in 2009. It was coordinated with the Australia's Future Tax System (AFTS) review, which was chaired by Dr Ken Henry when he was Secretary of the Department of the Treasury. The tax review was published in 2010. A common feature of the reviews, then, was that neither was chaired at arm's length from the Commonwealth Public Service. This stands in contrast with, for example, the work of the Asprey committee, which in 1975 recommended a broad-based federal indirect tax for Australia. AFTS panellists have likened their endeavours to Asprey's, but that committee was chaired by a judge of the New South Wales Court of Appeal.

That the chairs of the Harmer and Henry committees were at less than arm's length from the central bureaucracy has had predictable consequences. Positive ones include an impressive grasp of Australia's tax-transfer system, and moves towards rectifying anomalous tax expenditures originating in regional vested interests, such as the Fringe Benefits Tax treatment of cars, 'which may encourage individuals to travel unnecessary kilometres' (AFTS 2010b: 46). One negative consequence was disregard for the principle that investors, short of a national emergency, can expect that the rules will not change dramatically if investments outperform. Another was that growth-enhancing measures, such as cuts in business taxes, were to be in the 'short to medium term' and 'subject to economic and fiscal circumstances' (AFTS 2010a: 86), even while pension rises were unconditional and unaccompanied by serious new carrots and sticks for self-funded retirements.

The key Harmer recommendation resulted in an increase in the common value of the single rate of Age Pension and Disability Support Pension (DSP) from 25 per cent to 28 per cent of male total average weekly earnings. The key Henry recommendation was a Resource Super Profits Tax (RSPT) that would have initially taxed mining 'rents' at a headline rate of 36 per cent, on top of the pre-existing 30 per federal tax on mining profits. These recommendations represent two sides of the same coin: higher federal spending alongside higher federal taxes. The pension rise can be expected to reduce participation in the labour force, unless future governments can devise measures to restrict early retirement via the DSP. The proposed tax rise would discourage mining activity as miners considered their options to delay or abandon projects.

One giveaway of a process controlled by a central bureaucracy is lack of disclosure and transparency, and that is pervasive in the reports under discussion, particularly Henry's. For example, the Harmer-Henry recommendations envisaged that about 60 per cent of the median retiree's income in 2047 would come from the Age Pension rather than superannuation, notwithstanding the elapse of 55 years since the introduction of the Superannuation Guarantee (SG). So the SG is not viewed as supplanting the pension's longstanding role as the major source of retirement income even after the SG is fully mature. Yet this piece of information has to be gleaned from an undiscussed portion of a chart and lacks comment. Likewise, Henry says that his recommendations would be revenue neutral in 'steady state', yet he does not provide the information needed to check out the relevant calculations. Whereas he envisaged a drawn-out transition from the current 30 per cent rate of company tax to a 'steady-state' rate of 25 per cent, the RSPT was envisaged as coming on stream by the 2011-12 budget. In this way, Henry called for a tax reform that was strongly revenue-positive for several years--or possibly much longer than that--yet failed to put numbers on the expected departure from revenue neutrality.

Less predictably, Henry also sought to improve the tax-transfer position of people at the lower and upper extremes of the income distribution relative to people in the middle. The government rejected this particular set of recommendations, presumably on account of worries about its likely reception by the median voter--a propensity of governments that is sometimes described as Director's Law. The government also set out to water down the RSPT without totally abandoning it. The watered-down version is to be confined to coal and iron ore, and is projected in the forward estimates to yield about $5 billion per annum in 2012-13.

In mid 2011, then, winners in the short term out of the Harmer-Henry process appeared to include pensioners and Commonwealth public servants. (2) Losers in the short term appeared to include investors in mining companies and members of the workforces of Western Australia and Queensland. Losers in the long term could be a much wider group, as a consequence of the priority given by the Harmer-Henry process to increases in pensions and other public spending over measures to promote saving and investment.

2. Harmer on Pensions

The Harmer report was timely. In 2008, the total number of recipients of pensions from the Commonwealth was 3.3 million, compared to a population of 21.5 million. Hence the ratio of Australians as a whole to pensioners was 6.5 to 1. 'Pensioners' exclude recipients of Newstart, the Mature Age Allowance and assistance to lone parents not maintained by an ex partner. The Age Pension accounted for 2.1 million of the total number of pensioners. The DSP had 742,700 claimants. (3)

Harmer notes the large number of recipients of the DSP, which seeks to target people with an impairment that prevents them from working at least 15 hours per week in the two years following their health setback. Like the Age Pension, the DSP is indexed to the maximum of rises in male total average weekly earnings and rises in the Consumer Price Index (CPI). By contrast, Newstart and the Mature Age Allowance are indexed only to rises in the CPI. As a consequence, there is an ever-growing incentive to move out of those benefit categories and on to a DSP, which is of course more expensive to taxpayers. In 2008, the Newstart payment rate, to a person aged over 21 and with no children, stood at 79.9 per cent of the base rate of Age Pension paid to a single person. Accordingly, the phenomenon of benefit shopping goes a long way towards explaining the strong growth in DSP numbers.

Harmer's Finding 29 is that the DSP should 'more actively address questions of workforce participation' (Harmer 2009: xxi). It is hard to disagree. The devil lies in the detail, however, and Harmer confines himself to generalities such as a call for 'improved assessment procedures' (ibid.: 143). The United Kingdom has a similar program called Incapacity Benefit and has been similarly tentative in dealing with program growth. (4)

Another reason the Harmer review was timely is that for a decade or more there has been a largely unexamined drift away from the traditional Australian emphasis on means testing. The main arguments for and against targeting were covered in Mitchell et al. (1993). That article is a classic and can be read as a persuasive case for retaining our tradition of targeting. Mitchell et al. (1993) also noted the main argument against targeting: effective marginal tax rates (EMTRs) can be high. In 2008, if you moved from the Age Pension to paid work, not only did your benefit abate at the rate of 40 cents per dollar, but income in excess of the 'free' (tax threshold-plus-offsets) area was taxed at the rate of 15 cents in the dollar. In this way, your effective marginal tax rate reached 55 (= 40 + 15) cents per dollar of pre-tax non-benefit income. Targeted welfare can create poverty traps. Harmer updates calculations of EMTRs and makes recommendations for reducing them.

Against this line of argument, the defenders of Australia's tradition of targeting make two points. First, high EMTRs operate only over a limited range of incomes. Hence, as your non-benefits income rises, your benefits progressively cut out altogether. Once you are past the cut-off, EMTRs coincide with regular marginal tax rates, whereupon the problem of welfare-induced poverty traps becomes less severe. (5) Second, countries which have traditionally championed the universalist approach, such as the United Kingdom, Sweden and New Zealand, have recently been tightening up their criteria for benefits eligibility, in response to growing pressures on their national budgets. Universalism is expensive.

Harmer sheds light at various points on how far the pendulum has swung towards universalism in the Australian system. Chart 1 (Harmer 2009: 9), for example, suggests that someone who has had a full career at the 90th percentile of the earnings distribution will nevertheless be entitled to at least $10,000 per annum of Age Pension. On p. 126, Harmer notes that a partnered homeowner in 2008 could have assessable assets of $873,500 and still be eligible for a small amount of Age Pension, along with carded benefits. Yet he does not take a strong stand on the issue of universal versus targeted benefits. There is perhaps a tilt towards targeting, as in Finding 26, which says 'there would be capacity to tighten income test settings to limit the flow-on of [an] increase to pensioners with low to moderate reliance on the pension' (ibid.: 134). Overall, however, 'with the exception of the point at which the Seniors Concession Allowance is withdrawn, additional earnings increase the disposable income of the pensioner' (ibid.: 124). (6) In such ways, Harmer argues that targeting is working, at least to the extent of rendering EMTRs in excess of 100 per cent the exception rather than the rule.

Finding 28 recommends a deeming approach for accounts-based superannuation. This would simplify EMTRs by ensuring that owner-occupied housing was the only asset treated differently from other assets by the means tests. In particular, this 'extended deeming' would end the nightmarish complexity that used to come under the heading of the 'assessable amount' and is described by Harmer as 'frontloading' of the concessional treatment of income streams from superannuation. Among all the recommendations in the Harmer-Henry process for changes on grounds of simplicity, this is the most persuasive. On the other hand, implementing it would further reduce the tax-transfer efficiency of superannuation relative to owner-occupied housing.

Harmer notes: 'In consultations, some organisations and individuals raised concerns about the exemption of owner-occupied housing from the assets test' (ibid.: 140). He counters with a range of arguments, including one that owning your own home helps to manage 'longevity risk, which stretches the usual definition of that term. There are at least two arguments for toughening up the means tests on the family home, one old and the other new. The old argument is that allowing parents first to finance their retirements with the Age Pension and then to bequeath untaxed a multimillion-dollar home to their children is unfair both to low-wealth taxpayers and to parents bequeathing superannuation funded by employer contributions. My new argument is from efficiency: the resource boom is likely to require decades of more frequent moves interstate by some workers. The costs of selling the old family home and then buying a new one eat up about 8 per cent of the value of a house. In this way, it would help manage the upcoming era of more frequent housing transitions if people were less tempted by the tax-benefit system into premature house purchase than has historically been the case.

Ageing of the population naturally raises concerns about rising costs of pensions. Harmer sends a mixed message. He notes that the ratio of working people to people over the age of 65 has been projected to fall from 5.0 in 2007 to 2.4 in 2047. He estimates that social expenditures between 2007 and 2047 will require an increase of more than 23 per cent in government receipts, or 5.1 per cent of gross domestic product (GDP), assuming no increase in the maximum rate of pension. The SG will only reduce pension spending by 6 per cent. The participation rate in 2008 of men aged 55 to 59 is 77 per cent, compared to 90 per cent in the 1970s. Despite these estimates, Harmer is on the whole sanguine about the impact of population ageing on public finances: Australia's income support system is still relatively well placed to deal with demographic change' (ibid.: 13). The reason he gives is that Australia is not encumbered with a national superannuation scheme of the pay-as-you-go, defined-benefits variety, in contrast to the bulk of countries in the OECD.

Harmer's key term of reference was to make a recommendation about the full rate of Age Pension, and Finding 5 is the main one of his report. It is to increase the single Age Pension from (typically) 60.7 per cent of the rate paid to a couple to somewhere between 64 and 67 per cent. (7) Finding 1, which was to equalise the rates of Age Pension, DSP and Carer Payment, would see this increase flow through to DSP recipients. The government subsequently implemented these recommendations in the form of rises in the single Age Pension and the DSP from 25 per cent to 28 per cent of male total average weekly earnings.

Harmer wants to mitigate the expense of these measures. His last finding, numbered 30, calls for phased increases in the age of eligibility for the Age Pension, scheduled to equal 65 for males and females alike in 2014. He envisages an increase in the age of eligibility of between two and four years. Yet his own data on average ages at retirement, along with research by Hughes (2008), casts doubt on whether raising the age of eligibility would produce substantial savings. In more detail, Harmer finds that the average age at retirement in 2006-07 for those who retired in the previous five years was 61.5 for men and 59 for females. Similarly, Hughes found that Household Income and Labour Dynamics in Australia (HILDA) ages at retirement in 2002-03 were 61 for males and 57 for females. Hughes also showed that most men retired well before the eligibility age of 65, although attainment of that age did act to trigger a local peak in retirements. See Figure 1.


Figure 1 suggests that raising the male age of eligibility from 65 to (say) 67 would have only a small effect on pension outlays, for the simple reason that most male workers have already retired well before age 65.

Likewise, most women retired well before age 62, that being the female eligibility age in 2002, although there was a local peak in retirements in the neighbourhood of the Age Pension eligibility age for females in 2002--namely, 62. Figure 2 makes the point.


The life-cycle model with an extensive margin of labour supply can be used to make the point that lifting a retirement-conditioned pension relative to average earnings encourages people to retire earlier (Figure 3).


In Figure 3, the horizontal axis measures time elapsed since the start of working life. The maximum feasible span of working life is T. Two possible spans of working life with positive amounts of time spent in retirement are Ri and R2. The vertical axis shows accumulated assets. It also shows your reservation level of retirement assets--that is, the minimum amount you would need to be paid to induce you to retire immediately. You retire if and when actual assets hit reservation retirement assets, which depend positively on your maximum remaining span of working life and your net wage, and negatively on your disutility of work. Your initial wage earnings, net of taxes and retirement conditioned pensions, is Yi. A rise in retirement-conditioned pensions, as recommended by Harmer, cuts your net wage to Y2, cutting the opportunity cost of early retirement. (8)

3. Henry on Taxes

Treasury (2011) provides a useful snapshot of the national tax system. In 2008, Australia's ratio of tax to GDP was 27 per cent. This was below the OECD average of 35 per cent. Much of the difference can be explained by the fact that our mandated employer superannuation contribution of 9 per cent of wages is not classified as a tax instrument, whereas the compulsory employer and employee contributions that many countries levy to help pay for national defined-benefit superannuation schemes are classified as taxes. (9) In 2008, we raised 64 per cent of our revenue from direct taxation--that is, from a base consisting mainly of wages, salaries, payrolls and profits. This was above the OECD average of 60 per cent. The remaining 36 per cent of Australia's taxation revenue was derived from indirect taxation, including the Goods and Services Tax (GST); the OECD average was 37 per cent. In 2008, the Commonwealth raised 82 per cent of Australia's total tax revenue. This proportion of total taxation revenue attributed to the central government in Australia was the second highest among the countries that have a federal system of government. By contrast, Australia had the fourth lowest proportion for goods and services taxes. Taxes levied on superannuation funds raised around 2 per cent of Commonwealth revenue, or $12 billion in terms of 2007-08 dollars, even though only a small fraction of retirement incomes derives from superannuation. This disparity has arisen from the 'front-end' taxes on superannuation that Australia copied from New Zealand over two decades ago.

Household wealth data shed light on the incentives created by our tax-transfer system; see ABS (2007). In 2005-06, average gross household assets stood at $655,000 and household net worth stood at $563,000. Owner-occupied dwellings accounted for 44 per cent of average gross household assets and household contents averaged $51,000. About 20 per cent of households owned property other than the dwelling in which they lived. The value of this property averaged $91,000 over all households, accounting for 14 per cent of average gross household assets. By contrast, superannuation balances averaged $85,000 per household. Half of the 76 per cent of households with some superannuation had balances less than $44,000.

Like the Harmer review of pensions, the Henry review of taxes was timely. It was the first major effort at tax reform since the process a decade or so ago which saw the Ralph report on business taxes and the introduction of a 10 per cent GST. Henry listed the reasons why a fresh effort was due: population ageing, increased concern for the environment, increased economic integration with Asia and its associated resources boom, and continued international movement towards lower company taxes.

The trend to lower business taxes has taken place in tandem with refurbished arguments by economists for cutting capital income taxes. (10) Henry alludes to this literature: 'the review has drawn on the latest developments in economic theory' (AFTS 2010b: 15). This claim is broadly correct so far as taxation of capital income is concerned and the following passage is a neat summary of some points made by the recent literature: 'Taxes on savings income, including the taxation of inflationary gains, can discriminate against taxpayers who choose to defer consumption and save. The longer the person saves and reinvests, the greater the implicit tax on future consumption' (AFTS 2010b: 32). Henry overlooks the implication that the income-tax benchmark traditionally used for measuring tax expenditures should give some ground to a consumption-tax benchmark, for reasons of equity and efficiency alike.

Henry made 138 recommendations for tax reform. These are handily grouped and summarised in AFTS (2010a). Recommendations 1 to 25 concern personal taxation. Recommendation 1 was to concentrate revenue-raising on personal incomes, business incomes, economic rents and private consumption. This doctrine has an implication for measuring tax expenditures that was noted above; it strengthens the case for adopting elements of a consumption-tax benchmark. But Henry overlooks this case and the personal income tax effectively remains primus inter pares.

Recommendations 2 to 17 concern the personal income tax. Recommendation 2 harks back to a 1980s reform agenda which emphasised flattening the rate scale. Notably, in place of the existing first step in the rate scale, which sees incomes between zero and $6001 taxed at 15 per cent (disregarding offsets and the Medicare Levy), Henry proposes a tax-free threshold of $25,000. See Table 1.

Henry points out that this proposal would enable the abolition of a range of complex offset provisions, thereby promoting transparency and simplicity. Drawbacks of this proposal are twofold. First, it would entail giving up the nontrivial revenue currently raised from taxing wages at the current lowest marginal rate. In 2007-08, the 29 per cent of taxpayers who earned less than $30,000 in taxable income contributed 4 per cent of Commonwealth revenues. Second, standard static tax theory tells us that the deadweight loss caused by a tax rises with the square of the tax rate. For this reason, a 15 per cent marginal rate can be expected to impose a low deadweight cost. (11)

Recommendation 14 calls for a 40 per cent discount on tax payable on income from savings outside superannuation. The discount would not extend to dividends or rental income from non-residential properties. The non-taxation of owner-occupied housing remains unchanged. In this way, the income discount for capital gains and losses would fall from 50 to 40 per cent and the full deductibility for interest expenses relating to shares and rental property would be reduced to 40 per cent.

Under Recommendation 14, negative gearers would enjoy a less generous tax break than hitherto. That group accounts for 70 per cent of landlords. But Henry holds few terrors for them, if only because the recommendation is to be placed on hold until the current 'housing shortage' eases. Another giveaway of low priority is a lack of relevant modelling. Thus, Henry's numeric analysis of taxes paid by property investors assumes that rental property is held ungeared and then sold after seven years. It does not acknowledge the real options associated with negative gearing, as when a negative gearer converts her rented property into a principal residence in retirement, thereby avoiding capital gains tax. (12) Henry explicitly acknowledges a lack of modelling of interactions between the 40 per cent discount and extended deeming.

Recommendation 18 says 'the tax on superannuation contributions in the fund should be abolished' (AFTS 2010a). Instead, employer superannuation contributions should initially be treated as income in the hands of the individual and taxed at marginal personal income tax rates, and then receive a flat-rate refundable tax offset: 'The offset should be set so the majority of taxpayers do not pay more than 15 per cent tax on their contributions.' An effect of this recommendation would be to lift the compulsory contribution rate from its current effective rate of (1 - 0.15) x 9 = 7.65 per cent to the current headline rate of 9 per cent, where the figure of 0.15 refers to the current tax on employer contributions. Recommendation 19 says: 'The rate of tax on superannuation fund earnings should be halved to 7.5 per cent.' Also, 'the 7.5 per cent tax should also apply to capital gains (without a discount) and the earnings from assets supporting superannuation income streams'.

This proposal for progressive and primarily front-end taxation of superannuation is reminiscent of the superannuation surcharge that was levied between 1996 and 2005. The structure would be more progressive overall than the present flat-tax regime, but less progressive than the personal income tax. In particular, Henry floats the possibility of effective rates of zero, 15 and 30 per cent, corresponding to his two-part rate scale for the personal income tax (recall Table 1).

The existing headline earnings taxes are (i) 15 per cent on investment income from all superannuation balances other than those in drawdown mode and belonging to an individual aged over 60; and (ii) 10 per cent on realised capital gains on all superannuation balances other than those in drawdown mode and belonging to an individual aged over 60. The proposed earnings tax is a uniform one of 7.5 per cent. That this tax would fall on 'assets supporting superannuation income streams' is at odds with Henry's Term of Reference 5, which called on him to 'preserve tax-free payments for the over 60s' (AFTS 2010b: viii).

Young workers at the lower end of the income distribution would gain from Recommendation 18, as would those able to take advantage of new contributionsplitting opportunities. On the other hand, elderly people exposed to the new tax of 7.5 per cent on assets supporting income streams after retirement would lose out, as would fast-trackers on high wages. More than before, fast trackers would need to consider saving for retirement via negative gearing.

Recommendations 26 to 44 concern investment and entity taxation. Key ones are 26, that the 'structure of the company income tax system should be retained in its present form, at least in the short to medium term', and 27, that the 'company income tax rate should be reduced to 25 per cent over the short to medium term with the timing subject to economic and fiscal circumstances. There is something of St Augustine's plea in the timing of the proposed cut from 30 per cent to 25 per cent. The rationale is to move the company tax rate 'towards the lower end of the small to medium OECD economy average' (AFTS 2010b: xix).

Recommendations 45 to 50 concern land and resource taxes. These proved to be the most contentious of all. Here is the first one, quoted in full from AFTS (2010a: 89).

Recommendation 45: The current resource charging arrangements imposed on non-renewable resources by the Australian and State governments should be replaced by a uniform resource rent tax imposed and administered by the Australian government that:

a. is levied at a rate of 40 per cent, with that rate adjusted to offset any future change in the company income tax rate from 25 per cent, to achieve a combined statutory tax rate of 55 per cent;

b. is levied at a rate of 40 per cent, with that rate adjusted to offset any future change in the company income tax rate from 25 per cent, to achieve a combined statutory tax rate of 55 per cent;

c. applies to non-renewable resource (oil, gas and minerals) projects, except for lower value minerals for which it can be expected to generate no net benefits. Excepted minerals could continue to be subject to existing arrangements if appropriate;

d. measures rents as net income less an allowance for corporate capital, with the allowance rate set at the long-term Australian government bond rate;

e. requires a rent calculation for projects;

f. allows losses to be carried forward with interest or transferred to other commonly owned projects, with the tax value of residual losses refunded when a project is closed; and

g. is allowed as a deductible expense in the calculation of income tax, with loss refunds treated as assessable income.

AFTS (2010b: 235) sheds light on this six-part recommendation by explaining how the resource tax rate [t.sub.r] would interact with the company income tax [t.sub.c]. The objective is to 'keep the combined statutory tax rate on resource rents collected at the corporate level steady over time at 55 per cent', that is:

(1-[t.sub.c])[t.sub.r] + [t.sub.c] = 0.55, (1)

so the resource rent tax is given by

[t.sub.r] = 0.55-[t.sub.c]/1-[t.sub.c]. (2)

For example, the current company tax rate of 30 per cent would be supplemented by an RSPT of 35.7 per cent, and the proposed steady-state company tax rate of 25 per cent would be supplemented by an RSPT of 40 per cent, as outlined in Recommendation 45. The company tax would continue to apply to the conventional definition of profits involving depreciation schedules promulgated by the Australian Taxation Office. However, the RSPT would apply to a new definition of profits, described as 'resource rents', and viewed as proxies for the cash flows generated by mining businesses.

The design of the RSPT is loosely based on the cash-flow tax proposed in 1948 by the late E. Carey Brown. He assumed constant resource prices and concluded that a cash-flow tax which allows full expensing of investments would be nondistortionary. This result provided the justification for the hefty rate of RSPT, which would hit 55 per cent if ever the company tax were abolished. Hausman (2010) restates and updates Brown's ideas, and this exposition is instructive.13 Hausman notes that a cash-flow tax at rate t on the difference between mine revenue R, on the one hand, and the sum of current expenses z and net capital investment I, on the other, raises revenue T according to

T = [tau] (R - z - I), (3)

thereby allowing for full expensing of investment in the current year, along with reimbursement by the government of losses at the rate t. In place of the actual cash flows between the federal government and mining companies that would underpin a true Brown tax, Henry proposes an 'allowance rate set at the longterm Australian government bond rate' (AFTS 2010b: 231) multiplied by some estimate of enterprise value. Henry regards his proxy as retaining the essence of Brown's cash-flow approach, and makes observations similar to the following passage quoted by Hausman (2010: 4) from Brown's article: 'By paying the entrepreneur the tax on the asset's cost, the government would literally be a partner in the firm. It would make a capital contribution on new investment at the same rate in which it shared in the future net receipts of the enterprise.' The traditional net present value (NPV) of a project before tax is the expected discounted sum of the cash flows R - z - I. Since R - z - I = 0 if and only if [tau] x (R - z - I) = 0, the NPV of a marginal project--that is, one with NPV = 0--would not be affected by a Brown tax. So, imposing a Brown tax would not distort the decision about whether to go ahead with a project.

In reality, however, resource prices are volatile and this consideration overturns the neutrality of the Brown tax. The option value of delaying or abandoning prospecting for or exploiting a mineral deposit is substantial because of the valuable flexibility that enables a miner to slow down activity when price is temporarily low and speed up activity otherwise. The effects of price volatility on mine valuation and management have been rigorously analysed at least since Brennan and Schwartz (1985), which by mid 2011 had received 1446 citations. Brennan and Schwartz applied elements of the successful Black-Scholes theory of valuing financial options to valuing mining projects. (14) Hausman (2010) follows this literature in pointing out that we need a modified rule for projects characterised by volatile output prices and sunk investment costs: go ahead today with the project if and only if [NPV.sup.full] = 0, where [NPV.sub.full] includes the net present value of real options embedded in the project that would be killed if the project were to go ahead today. These options include the value of waiting for prices to rise before proceeding. According to Hausman (2010: 7), the 'common finding' is that total embedded option values are '2-3.4 times the size' of capital investment. Yet the proposed RSPT leaves these option values out of account, as can be seen from the right-hand side of equation (3). So the RSPT would be unlikely to possess the desirable efficiency properties claimed by Henry.

The RSPT also has worrying implications for existing mining projects that involve no further investment. Hausman (2010) points out that the effective rate of RSPT would exceed the nominal steady-state rate, namely 40 per cent, because the government would never have made its initial 40 per cent investments in projects. In this way and others, the RSPT would be highly retrospective. Of course, all new taxes have elements of retrospectivity, but the sheer magnitude of the proposed RSPT heightens concerns with fairness and would amount to a big change in the customary relationship between business and government in Australia. For an initiative of comparable magnitude, you have to go all the way back to the proposed nationalisation of commercial banks by the Chifley government in 1947.

Recommendations 51 to 54 concern land tax and conveyance stamp duty. They canvass options for replacing stamp duties with taxes on unimproved land values. Like his proposed 40 per cent discount in incomes from assets held outside superannuation, Henry gave the impression that reforms along these lines are on the back burner. Recommendations 55 to 57 are a miscellany. Number 57 says: 'State payroll taxes should eventually be replaced with revenue from more efficient broad-based taxes that capture the value-add of labour.' This smacks of an ambit claim and there is no modelling to back up the claim of superior efficiency.

Recommendations 58 to 81 are about 'enhancing social and market outcomes, including taxes to improve the environment. In particular, Recommendations 61 to 68 concern road transport taxes. The most interesting of these is 61, which says 'governments should analyse the potential network-wide benefits and costs of introducing variable congestion pricing on existing tolled roads (or lanes), and consider extending existing technology across heavily congested parts of the road network. Henry goes on to point out that 'new technologies may further enable wider application of road pricing if proven cost-effective. In general, congestion charges should apply to all registered vehicles using congested roads. This is an idea whose time has come.

Recommendations 69 and 70 seek to improve housing affordability. For example, state governments could cut developer charges. Accordingly, these recommendations rely on acquiescence by state governments and are in the nature of an ambit claim. Recommendations 71 to 78 are for the taxation of alcohol, tobacco and gambling. Recommendation 71, for example, says that all alcoholic beverages should be taxed on a volumetric basis, and that the rate of alcohol tax should be based on evidence of the negative externality created by alcohol. This package of prescriptions is broadly in line with those of Ramsey and Pigou, although premium alcoholic beverages may have an element of insensitivity to price that justifies a higher tax on them.

Recommendations 79 to 81 represent a residual category, covering insurance, luxury cars and user charges. That 'insurance products should be treated like most other services consumed within Australia and be subject to only one broadbased tax on consumption' may not go far enough to ensure a consumption-tax treatment of insurance purchase, which is a form of saving.

Recommendations 82 to 110 concern the transfer system and therefore overlap with Harmer's recommendations. They tend to be broader than Harmer's because they span non-pension recipients of Commonwealth income support. Recommendation 85, for example, says that recipients of income support for parents should be required to look for part-time work once their youngest child turns four, which is on the low side of the ages that have traditionally been canvassed in the Australian debate. Recommendations 111 to 138 concern institutions, governance and administration of federal and state taxes, and are not a priority for economic analysis.

4. The Future

There is a lot to like at the level of detail in the Harmer-Henry package. But future efforts to reform our tax-transfer system will need a different big picture.

Future endeavours to reform pensions will need to address the tough questions. These include more involvement of government doctors in assessing claims for DSPs, more check-ups in the years after a DSP has been initially approved, indexation of pension benefits to the Consumer Price Index alone rather than to the maximum of rises in prices and wages, acknowledgement that refraining from increasing Age Pensions and DSPs relative to average earnings helps to deter premature retirements, and recognition that a household with several millions in housing wealth together with close to a million in superannuation assets is not obviously a deserving recipient of a part Age Pension.

Future endeavours to reform taxes should begin with an attempt to determine which programs are better delivered at a state and local level rather than federally. For example, before World War II, Australia followed North America in delivering secondary and tertiary education through the state level of government. After the war, however, society's growing need for investments in human capital was increasingly met by federally funded initiatives, partly because Commonwealth governments and the High Courts had acted to remove major tax bases from the states, and partly because of the ongoing cultural influence of educators and administrators from the United Kingdom--which is, of course, a unitary state. (15)

Indicators of university performance include the first preferences of international students, the country of origin of publications in leading journals, and the country of origin of patents and copyrights. These suggest that North American institutions have come to represent international best practice in tertiary education. We should now follow the North American model rather than the British model. This would entail less university funding by way of federal grants and cross-subsidisation from international students seeking permanent residence. There would be more university and school funding by way of state grants, more freedom to charge realistic fees to local students, and more tax breaks for local students and their families who are paying their own way, along with more tax breaks for philanthropy extended to educational institutions. The case for educational tax breaks is of a piece with the case for cutting taxes on income from physical capital. In both instances, moving towards a consumption-tax benchmark would over time lift the stocks of physical and human capital, encouraging socially optimal levels of saving and driving up wages.

Students and state authorities would come to have more influence on university and school programs, while the federal authorities would have less. Competition between educational institutions would more closely resemble the kind of competition between businesses upon which we rely for value for money in markets other than educational ones. Universities and schools would be reconnected with their local communities. Equity could be protected by measures such as reviving our traditional means-tested Commonwealth scholarships.

Such a transfer of functions back to the states could be funded in part by adding a new slice to the GST. On top of the existing 10 per cent rate, there could be a rate initially set at 5 per cent, and allowed subsequently to be varied across states, by the decisions of state governments. The Commonwealth could continue to collect all GST receipts, but would reimburse precisely 100 per cent of the second slice in accordance with the rate set by the state in question. This reform would facilitate the proper functioning of 'Tiebout sorting', whereby households move to state jurisdictions that best suit family tradeoffs between disposable income and access to public services such as public education. (16) In the United States, for example, education-conscious households often reside in high-tax states that support an 'Ivy-public' university.

Superannuation too is an area where the United States and Canada have instituted policies worth copying, if only in modified form. We could give workers a choice between either or both of two kinds of super account. One would be taxed and regulated under the current arrangements. The other would be taxed only in retirement and at the marginal rate of the retiree. The new accounts would be reserved for the purchase of lifetime annuities. This element of compulsion would help rectify the demand-side problems that have stunted the development of a deep and fair market for lifetime annuities in Australia. The equity of the superannuation system would improve as people retiring on modest balances would on a lifetime basis pay little tax on superannuation held in these accounts. Moreover, in contrast to Henry's proposed progressivity via a two-step tax on employer contributions, you would avoid the risk of paying hefty superannuation taxes during the accumulation stage and then retiring on a meagre income in the wake of a market crash on the cusp of retirement. (17)

Exposure to growth assets within the new accounts, once annuitised, would be capped at 50 per cent, helping to protect both the budget and self-funded retirees. Harmer noted that 'Age Pension applications in December 2008 were around 50 per cent higher than the number recorded in October of the same year' (2009: 15). However, neither Harmer nor Henry joined the dots between this extraordinary shock to public and private budgets alike and our policy not only to eschew restrictions on asset allocations in superannuation accounts, but also to levy a higher effective tax rate on earnings from interest-bearing securities than on earnings from growth assets. Indeed, financial planners highlight this tax distortion in our superannuation system as part of their sales pitch for persuading elderly investors to reweight towards growth assets. (18)

Likewise, Harmer and Henry were sanguine about the prospect of the median male worker relying on the Age Pension for 60 per cent of his retirement income even after the system had had 55 years to mature. Future tax reviews should instead back the proposed increase in the SG from 9 to 12 per cent of wages. This is a more promising way to avoid squandering the resource boom than setting up a sovereign wealth fund. If the current re-regulation of the labour market sees the effective incidence of a rise in the SG fall increasingly on employers rather than employees (thereby creating unemployment), then we should make employees directly liable for the extra 3 per cent contribution. Another advantage of a 3 per cent contribution falling on employees is that it could cut in at a different age from employer contributions--for example, at 40 years of age. In this way, the time-path of total compulsory contributions would look more like the way a farsighted person would schedule their lifetime contributions in the absence of either compulsion or state-provided retirement income. It would make our system look more like Switzerland's, which is close to international best practice in mandated mass superannuation.

Future initiatives to cut business taxes should avoid stipulating ill-defined hurdles that need to be overcome before the cuts can be implemented. No such hurdles impeded the Harmer-Henry recommendations to lift the DSP and the single rate of Age Pension.

Finally, future tax reforms should moderate Henry's proposed tax on 'resource rents. Investors have a right, short of a national emergency, to expect that the rules will not change dramatically if investments outperform. Australia has broadly respected this principle for two centuries and it has served us well. Revisiting the tax on miners looks likely, if only because of the arbitrarily narrow base--coal and iron ore--of the Minerals Resource Rent Tax, successor to the RSPT. Before Treasury goes back to the drawing board, it should beef up its capabilities in modern real-options techniques for valuing and managing mines. Henry's adherence to the dated ideas of E. Carey Brown evidently led him to overlook the options possessed by miners to delay or abandon projects. In this way, he ended up exaggerating both the efficiency and the likely revenue yield of his proposed RSPT.


Atkeson, A., Chari, V. and Kehoe, P. (1999) 'Taxing capital income: A bad idea, Federal Reserve Bank of Minneapolis Quarterly Review, 23(3), pp. 3-17.

Australia's Future Tax System (AFTS) (2010a) Report to the Treasurer, December 2009: Part One--Overview, Commonwealth of Australia, Canberra.

Australia's Future Tax System (AFTS) (2010b) Report to the Treasurer, December 2009: Part Two--Detailed Analysis, Part 1 of 2, Commonwealth of Australia, Canberra.

Australia's Future Tax System (AFTS) (2010c) Report to the Treasurer, December 2009: Part Two--Detailed Analysis, Part 2 of 2, Commonwealth of Australia, Canberra.

Australian Bureau of Statistics (ABS) (2007) Household Wealth and Wealth Distribution, Australia, 2005-06, Cat. No. 6554.0, ABS, Canberra.

Bateman, H. and Kingston, G. (2010a) 'Tax and super--unfinished business, JASSA: The Finsia Journal of Applied Finance, December, pp. 49-54.

Bateman, H. and Kingston, G. (2010b) 'Henry and super and saving, Australian Economic Review, 43(4), pp. 437-448.

Brennan, M. and Schwarz, E. (1985) 'Evaluating natural resource investments, Journal of Business, 58(2), pp. 135-157.

Brown, E. (1948) 'Business-income taxation and investment incentives' in L. Metzler (ed.) Income, Employment and Public Policy: Essays in Honor of Alvin H. Hansen, Norton, New York, pp. 300-316.

Harmer, J. (2009) Pension Review Report, Commonwealth of Australia, Canberra.

Hausman, J. (2010) Analysis of the taxation of rent: Mineral industries in Australia, available: [accessed 5 July 2011].

Hughes, K. (2008) Retirement Income Policy in Australia, Fourth year Honours thesis, Department of Economics, University of New South Wales, Sydney.

Judd, K. (2001) 'The impact of tax reform in modern dynamic economies' in K. Hassett and R. Hubbard (eds) Transition Costs of Fundamental Tax Reform, AEI Press, Washington, DC, pp. 5-53.

Keane, M. (2010) Labor supply and taxes: A survey, Working Paper 160, School of Finance and Economics, University of Technology, Sydney.

Kingston, G. (2000) 'Efficient timing of retirement, Review of Economic Dynamics, 3(4), pp. 831-840.

Kingston, G. (2006) 'Choice of tax regime for superannuation contributors, Australian Accounting Review, 16(3), pp. 41-46.

Kingston, G. (2009) 'Financial plans for baby boomers: How much risk?, Economic Papers, 28(2), pp. 65-74.

Mitchell, D., Harding, A. and Gruen, F. (1994) 'Targeting welfare, Economic Record, 70(10), pp. 315-340.

Pindyck, R. (1991) 'Investment, irreversibility and uncertainty, Journal of Economic Literature, 29(3), pp. 1110-1148.

Schwartz, E. (1997) 'The stochastic behaviour of commodity prices: Implications for valuation and hedging, Journal of Finance, 52(3), pp. 923-973.

Treasury (2011) Pocket Guide to the Australian Taxation System, Australian Treasury, Canberra.


(1.) I would like to thank two referees for helpful comments and the Australian Research Council for financial assistance via DP0877219.

(2.) More precisely, public servants will be winners to the extent that they want to manage bigger budgets.

(3.) At the time of writing, this figure had risen to 815,000.

(4.) One indication of the degree of similarity is the number of benefit recipients relative to the size of the labour force. Australian Bureau of Statistics data suggest that 6.8 per cent of the Australian labour force currently receives the DSP, whereas Wikipedia suggests that 8.5 per cent of the labour force in the United Kingdom receives Incapacity Benefit. Allowing for the lower level of economic activity in the United Kingdom at the time of writing, these percentages are close.

(5.) This traditional view needs to be tempered by new arguments that there are pervasive downward biases in estimates of the elasticity of labour supply, even if attention is confined to the intensive margin. If these arguments are valid, then the welfare costs of tax-transfer policies are greater than has traditionally been supposed. See, for example, Keane (2010).

(6.) In the case of a single pensioner, that point cuts in at a private income of $49,400 per annum and generates an EMTR of 120 per cent.

(7.) Harmer pays tribute to the pioneering research of Dr Bruce Bradbury of the University of New South Wales, who revitalised Australian research on equivalence scales. However, Bradbury apparently recommended that the single rate of pension be set at 68.5 per cent of the couple rate.

(8.) Figure 3's portrayal of the life-cycle model with an extensive margin of labour supply incorporates a number of simplifying assumptions. These include a zero-one work-retire decision, log utility from consumption, a zero rate of time preference in the relevant range, a constant net wage in the relevant range, constant disutility of work in the relevant range, no uncertainty, and no inherited assets or bequest. For details, see Kingston (2000).

(9.) There is, of course, a respectable case for not classifying mandated superannuation as a tax.

(10.) So far as the small open economy is concerned, the contribution by Atkeson et al. (1999) has perhaps been the most influential in academe. Note that the case for cutting the rate of tax on capital was originally made in a closedeconomy setting--see, for example, Judd (2001).

(11.) The new view of capital income taxes makes a similar point: the deadweight cost of taxing capital income rises nonlinearly with the delay between saving some amount and then consuming it.

(12.) For details, see Bateman and Kingston (2010b) and Kingston (2006).

(13.) Hausman's analysis was 'funded by BHP', but his credentials as a Bates Prize winner and member of MIT's economics faculty have weight. (The late E. Carey Brown was also a member of MIT's faculty.) Moreover, it would be naive to regard the Harmer-Henry process as a disinterested party in the debate on tax reform.

(14.) See Schwartz (1997) for an update of this line of research and see Pindyck (1991) for a primer on company valuation and capital budgeting that takes real options into account.

(15.) The United Kingdom has recently launched decentralising initiatives in education--namely, permission to start so-called free schools and to lift university fees (the United Kingdom cap on fees has been raised to 9000 [pounds sterling] per annum).

(16.) Judd (2001) promotes Tiebout sorting in the case of the United States.

(17.) For details, see Bateman and Kingston (2010a, 2010b).

(18.) For details, see Kingston (2009).

About the Author

>> Geoffrey Kingston is a Professor in the Department of Economics at Macquarie University. His research interests include superannuation, exchange rates, interest rates and tax policy. He can be contacted at

Geoffrey Kingston *

* Department of Economics, Macquarie University
Table 1: Current and proposed rate scales

 Current Proposed
Taxable income rate % rate %

$0-$6000 Nil Nil
$6001-$25,000 15 Nil
$25,001-$37,000 15 35
$37,001-$80,000 30 35
$80,001-$180,000 37 35
>$180,000 45 45

Source: Treasury (2011:13); ATFS (2010c: 22).

Note: The current and proposed rate scales are not strictly
comparable because Henry proposes to abolish the
Medicare Levy and the Low Income Tax Offset.
COPYRIGHT 2011 Centre for Applied Economic Research and Industrial Relations Research Centre
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kingston, Geoffrey
Publication:Economic and Labour Relations Review
Date:Nov 1, 2011
Previous Article:Pension adequacy and the pension review.
Next Article:Population ageing and tax reform in a dual welfare state.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters