Back at the crossroads: the slippery fish of Australian retirement income policy.

Introduction

To obviate the financial losses that would otherwise accompany leaving work in old age, most countries have developed retirement income systems. These systems employ a limited range of governmental, quasi-governmental and private mechanisms which, through institutionalizing 'stable' sources of income support in later life, seek to protect workers against the financial risk of retirement (Kingson & Williamson 2001).

Australia has similarly sought to protect its workers from the risk of a loss of earned income due to retirement. However, the system that existed for the first seven or eight decades following federation in 1901 was really a non-system. A crossroad was reached in the early 1980s when the belated recognition of this situation precipitated a flurry of reforms. These reforms were driven by an emergent yet, ultimately, reasonably clear set of goals that sought to transform the non-system into a relatively coherent set of arrangements that, in the future, would provide protection against the loss of earned income. While some elements of the new arrangements were the subject of criticism, particularly concerning their inequity, for a couple of decades at least it seemed that the prospects were quite good of broadly achieving a new retirement income system that, when mature, would realize the goals of the reform process.

Today, however, Australia's retirement income arrangements are once again at the crossroads. Recent years have witnessed a number of changes which have effectively created a system that is different in a number of significant respects from that sought by the reformers of two or so decades ago. Further, the inequities associated with contemporary arrangements have become palpable. What was supposed to be a retirement income system has morphed into something different, a system whose primary function is no longer to simply protect workers from the loss of earned income arising from retirement. The 'zenith' of this new situation was reached in the May 2006 Federal Government Budget in which major changes in superannuation were announced.

The purpose of this paper is to examine Australia's contemporary so-called retirement income system and identify some of the challenges it raises for policymakers. In order to understand how we arrived at where we now find ourselves, the paper begins by overviewing Australia's retirement income arrangement prior to the late 1970s-early 1980s. It then turns to a consideration of the goals of the 'new' retirement income system that began to emerge in the 1980s and the sense in which they represented a 'grand departure' from the non-system that was previously in place. How these major goals were pursued is the subject of the third section. In the fourth section the paper describes the 'state' of Australia's retirement income system at the dawn of the 21st century. This is followed by a consideration of some of the more contemporary changes, changes which reflect renewed policy sketchiness in the retirement income policy field, viz., the emergence of a new pillar of retirement income policy and the strengthening and reshaping of an existing one. The penultimate section looks closely at the changes in superannuation announced in the 2006 Federal Budget and explores some of their implications. The paper concludes with an attempt to articulate the contemporary nature of Australia's retirement income system and the implications of some of its major characteristics.

An underlying theme of this paper is that retirement income policy in Australia, like policy in other domains, is dynamic. At times its purpose and direction have been elusive or sketchy and, like a slippery fish, quite hard for the policy analyst to grasp. At other times, relative clarity has prevailed. It is the contention of this paper that as we approach the end of the first decade of the twenty-first century, retirement income policy in Australia has once again become hard to grasp. Unless Australian policymakers make clear the nature and desired outcomes for Australia's retirement income system it will be very difficult indeed to reach the desired policy destination.

Prior to the 1980s: Australia's Retirement Income 'Non-System'

A descriptor frequently used by students of the welfare state is the notion of the 'pillar'--who pays for and provides the range of cash transfers and social benefits that comprise the non-market income of individuals and households at different life stages (Rein 1999: 2). Retirement income arrangements in Australia have generally been thought of as being comprised of three pillars--the market pillar of occupational superannuation, the public pillar of the Age Pension and a third pillar of voluntary personal savings (including owner-occupied housing). Indeed, Australia's retirement income system has closely resembled the three-pillar model described in the World Bank's (1994) report, Averting the Old Age Crisis.

The Market Pillar

Prior to the 1980s Australia's occupational superannuation schemes played a minor role in providing retirement income despite a history that stretches back to the mid-nineteenth century. Prior to World War 2 coverage was largely confined to male white-collar public servants and employees of financial institutions and large manufacturing concerns. Despite a post-war expansion in coverage, it stood at just a third of the workforce by the early 1970s. A decade later coverage had reached only 45 per cent (Borowski, Schulz & Whiteford 1987). These figures belie the even narrower coverage of female workers. As recently as 1983 it was only 23 per cent, with only 11 per cent of women in the private sector workforce being covered (ABS 1998).

The limited role played by, typically, defined benefit occupational superannuation schemes in providing a retirement income was attributable to: (1) this narrow coverage, (2) the absence of preservation requirements that resulted in superannuation benefits being paid not only at retirement but also at any time a worker changed jobs and (3) the payment of benefits in lump sum form, a payment that was rarely used to purchase an annuity, (private pension income stream). The latter was a result of a taxation regime in which lump sums were very lightly taxed while income streams were fully taxed. Thus, the market pillar functioned as something more akin to a concessionally taxed severance pay device than a source of retirement income.

The Public Pillar

Given this situation, it was the public pillar of the Age Pension that served as the primary source of income in retirement for older people. Legislated by the Deakin Government in 1908 and introduced in 1909, this means-tested provision was envisaged as providing a minimum subsistence income that was supplementary to any personal savings (the third pillar) that older people may have accumulated themselves. As a result of progressive liberalizations in eligibility criteria, five decades later 70 per cent of people qualifying on the grounds of age received the Age Pension-up from 32 per cent in 1920.

Further easing in eligibility continued through to 1976 and culminated in the abolition of the assets test (i.e., the capital) component of the means test. In consequence, by mid-1978 three-quarters of the population of pensionable age received the Age Pension. If one adds Service Pensions paid to veterans, then the ostensibly selective public pillar of Australia's retirement income system had effectively become a demogrant covering about 90 per cent of people of eligible age (Department of Social Security 1984; Whiteford & Bond 2000). In 1978, however, some tightening of eligibility began when future pension increases for those aged over 70 were made subject to an income test. This change heralded the beginnings of a marked shift from what had become a widely held view that the Age Pension was something to which all older Australians were entitled.

In sum, then, as Australia entered the 1980s it had a host of occupational superannuation schemes that were funded by expensive (to the Government in terms of foregone revenue)) tax-deductible employer contributions but which played only a minor role in providing a retirement income. It also had the Age Pension which, although designed to assist the needy elderly, was being received by almost all Australians who were age eligible at great cost to the public purse.

The Goals of Australia's 'New' Retirement Income System as Grand Departures

Given this situation one could be forgiven for describing retirement income policy in Australia as a slippery fish--something quite difficult to grasp and comprised of pillars whose roles and goals were elusive. This situation required a reshaping of the system, a process that began in earnest with the election of a Labor Government under Prime Minister Bob Hawke in 1983. The main architect of this reshaping was the Treasurer (and, from late 1991 until 1996, Hawke's successor as Prime Minister), Paul Keating.

As this reshaping process unfolded, the following major goals for Australia's retirement income system emerged:

1. Reducing reliance on the Age Pension through tightening eligibility;

2. Enhancing the role played by the market pillar in providing retirement income through expanding occupational superannuation coverage;

3. Providing equitable superannuation coverage and, thus, employer-provided benefits for all employees

4. Improving the level of superannuation coverage with a view to enhancing the adequacy of future retirees' incomes;

5. Containing the costs to the public purse of superannuation tax concessions;

6. Enhancing the role played by superannuation as a retirement income stream; and

7. Increasing household and national savings and, in turn, investment.

These goals represented a grand departure from the non-system that was in place through to the early 1980s.

The Age Pension

The near-universal availability of the Age Pension resulted in it becoming entrenched in the public imagination as a social right based on the individual's social contribution as a worker and taxpayer. The 1978 change (and, indeed, the importance of the retirement income policy domain thereafter) broadly coincided with the (somewhat belated) public recognition of population ageing in Australia and its long-term costs (Howe 1981). This, plus budgetary pressures, led to some tightening of eligibility in 1983 and, in 1985, the reintroduction of an assets test. These and subsequent tightenings led to a decline in Age Pension coverage. While these declines were small they nevertheless reflected a significant shift in the construction of the Age Pension from a demogrant to an increasingly selective provision--a return to the residual role for the Age Pension envisaged by the Deakin Government.

Occupational Superannuation

The goals for superannuation were a grand departure in two respects. First, they represented a real attempt to articulate a role for superannuation. (1) Second, they reflected an acknowledgement by government that it was now legitimate to do what had previously been illegitimate, namely, to intervene with regard to middle- and high-income earners with a view to ensuring higher levels of earnings replacement in retirement, even though doing so would contribute to perpetuating into retirement the income inequalities that governments usually try to mitigate (Forward 1983).

Pursuing the Goals

Expanding and Improving Superannuation Coverage and Providing Equitable Coverage

The pursuit of the major goal of reducing reliance on the Age Pension involved efforts not only to tighten eligibility but also to expand and improve occupational superannuation coverage. The expansion in coverage was also intended to address the inequity of narrow coverage (the third goal). To the extent that occupational superannuation coverage grew, improved and, over time, matured, the need for the Age Pension would be substantially reduced. The strategy for expanding and improving coverage was initially a consensual one but ultimately demanded compulsion.

A distinguishing feature of pension reform observed 'with uncanny regularity' around the world has been the utilization of the corporatist social contract involving organized labour and employers' associations as the political mechanism for redesigning pension policies (Myles & Pierson 2001: 306). In Australia, however, the social contact was between organized labour (the Australian Council of Trade Unions (ACTU)) and the Federal Government.

The anti-inflationary Prices and Incomes Accord entered into between the ACTU and the Australian Labor Party on the eve of the 1983 federal election 'constrained' the ACTU to focus its wage claims on non-wage benefits and working conditions (the 'social wage'). In the 1986 national wage case it sought a 3 per cent wage equivalent and employer-paid superannuation contribution for employees working under national awards. The Conciliation and Arbitration Commission endorsed employer-employee negotiations for either new or improved superannuation based on 'award superannuation' (the 3 per cent wage equivalent). Under the Occupational Superannuation Standards Act (OSSA) of 1987 employer contributions deposited in individual workers' accounts held by superannuation funds had to be portable and preserved until age 55. (2) Award superannuation resulted in coverage increasing to 79 per cent of all employees by the end of the 1980s (APRA 2007: 3).

Award superannuation, however, excluded workers not covered by awards, including a sizeable minority of private sector employees. Further, some employers either resisted negotiations over award superannuation or failed to comply with Commission-approved consent agreements. And difficulty was experienced in policing employer payments of contributions. In the event, the government announced the introduction of mandatory superannuation coverage at the 1991 Federal Budget. (3)

Under the 1992 Superannuation Guarantee (SG) Charge Act all employers not already doing so were required, as of 1 July, to pay a minimum of 3 per cent of the employee's salary into a superannuation fund. In recognition of award superannuation being too small to provide a significant improvement in retirement income, higher levels of mandatory tax-deductible employer contributions were progressively phased in over the next decade and reached 9 per cent (4) in 2002. Mandatory contributions have remained at this level ever since.

As a result of the SG, today about 90 per cent of all workers almost all full-time workers and about three-quarters of part-time workers--have some superannuation coverage (Stanford 2003, ASFA 2007). Since 1990, self-employed and uncovered employees have been able to make tax-deductible contributions to superannuation funds within limits which, until very recently, were age-based. However, some coverage inequities persist. Thus, workers in low-wage employment remain uncovered. (5) Further, the lowest superannuation coverage is found in industries characterized by a high proportion of casual and/or part-time and female employees, e.g., retail trade, restaurants, etc. (ABS 2005). With the increasing trend to casual and part-time employment, a sizable number of people holding multiple part-time jobs are deriving no benefit from the SG with major consequences for their future retirement income.

While unquestionably enhancing the role of the private pillar, the expansion in coverage has been accompanied by other, less sanguine developments. The most notable has been the heightened financial risk borne by members of superannuation funds due to the marked shift from defined benefit to defined contribution or accumulation funds.

Prior to award superannuation, most fund members were in employer-sponsored defined benefit plans. However, the funds that were established to accept award and, subsequently, SG contributions were of the accumulation variety. So great has been the shift from defined benefit to defined contribution/accumulation funds that, by 2005-06, 97 per cent of members were either in accumulation funds or funds that provided a mix of accumulation and defined benefits (APRA 2007: 5). The risks borne by members of accumulation funds include fraud and mismanagement of funds, inflation, the longevity risk (outliving accumulated benefits) and the investment risk (fluctuating investment returns). (6) The latter risk is illustrated by the fall in the value of superannuation fund balances arising from the sharp losses experienced by equity markets in mid-2007 (Patten 2007) (losses which were soon reversed following a robust market rebound) and the even sharper and more enduring losses experienced since January 2008 in the wake of the US sub-prime ('low-doc' in Australian parlance) mortgage market crisis.

Enhancing Retirement Income Adequacy

The reform process that began in the 1980s was accompanied by a shift in the construction of what constitutes an adequate retirement income. Until relatively recently, the notion that an adequate retirement income is one that allows retirees to maintain their earnings-related accustomed pre-retirement living standards enjoyed very little currency in Australia (Borowski 1984; Borowski, Schulz & Whitefoord 1987). By the early 1990s, however, this alternate notion of retirement income adequacy had begun to take hold (Commonwealth of Australia 1992). It is now broadly accepted that for a worker on average earnings an earnings replacement rate in the range of 60-65 per cent is required to maintain accustomed pre-retirement livings standards (Commonwealth of Australia 2002). This earnings replacement rate decreases with higher pre-retirement earnings and increases with lower pre-retirement earnings.

However, a SG contribution of 9 over cent over a working lifetime by itself is unlikely to yield an adequate retirement income. Hence, many commentators (e.g., Warren 2004), former Prime Minister Keating (Hughes 2005) and stakeholders such as the ACTU and the superannuation industry, have advocated higher levels of superannuation contributions funded through additional employer, employee and/or government contributions. (7) While there have been no indications from either of the major political parties that higher contribution levels will be mandated any time soon, the Government has taken some ad hoc measures to enhance retirement income adequacy (see below).

Containing the Cost of Tax Concessions

A further goal of the reform process was to contain the public subsidy associated with the tax concessions (or 'tax expenditures') enjoyed by superannuation schemes. As noted earlier, prior to the 1980s employer superannuation contributions were tax deductible and lump sums were lightly taxed. Nevertheless, superannuation played only a minor role in providing a retirement income. The poor integration between the tax-funded Age Pension and tax-subsidized superannuation also meant that the receipt of a lump sum did not necessarily obviate Age Pension eligibility, a situation colloquially known as 'double dipping'.

In consequence of this, a number of superannuation taxes were introduced in a series of legislative changes beginning in 1984. (And in order to cap the level of tax-advantaged benefits the government considered reasonable for retirees to receive, it introduced Reasonable Benefits Limits (RBLs) in 1994.) Suffice it to note that by 1996 superannuation was subject to a unique and complex tax regime. Some elements of this regime were highly inequitable, a matter to which we return below.

The tax regime comprised a contributions or entry tax, a tax on superannuation fund earnings, an exit tax and a surcharge for high-income earners. These taxes certainly reined in the public cost of superannuation. Indeed, they generated enormous revenue for the government equivalent to just over 3 per cent of all federal government revenues (ASFA 2004). They also impacted negatively on the level of superannuation savings that would otherwise have been accumulated and thus had the effect of undermining one of the other goals of the reform process, viz., enhancing the adequacy of retirement incomes provided by superannuation. In somewhat grudging recognition of the impact of these taxes on superannuation savings levels, the surcharge was abolished from 1 July 2005. Nevertheless, superannuation taxes continued to yield very considerable revenues.

Superannuation as a Retirement Income Stream

Changes in the tax treatment of superannuation were not only designed to reduce public subvention of superannuation. They also sought to redress the situation of the more generous tax treatment of lump sums relative to annuities and thereby make the latter more attractive. Such measures as changes in the tax treatment of lump sums, a two-tiered RBL structure, a 15 per cent tax offset available to those aged 55 and over who drew income from an annuity and the generous treatment of the capital component of complying annuities under the Age Pension's assets test all sought to encourage the purchase of an income stream. Superannuation preserved to retirement and then used to purchase an annuity would also obviate the need for an Age Pension.

Despite the tax and other incentives, the government's efforts to encourage retirees to purchase annuities with their lump sums met with only limited success. As recently as 2003, three-quarters of all retirees continued to opt for a lump sum payment (Commonwealth of Australia 2003).

Australia's Retirement Income System at the Dawn of the 21st Century

After more than two decades of reform Australia entered the 21st century with a retirement income system which could point to some major achievements. There has certainly been an impressive expansion in superannuation coverage. Today 93.2 per cent of male employees and 91.8 per cent of female employees are covered (ABS 2007) although the challenge remains of extending coverage to the small proportion of the workforce that remains uncovered. Further, major regulatory reforms in response to the growth in defined contribution funds have gone a long way in protecting workers' superannuation savings. (8)

The SG regime has also resulted in higher levels of savings in superannuation funds than would otherwise have been the case. At the end of 2007 consolidated assets of superannuation funds totalled $802.4 billion (ABS 2007c) indicating that the seventh objective referred to above is well on the road to being realized. Nevertheless, the Age (and Service) Pension continues to be the major source of retirement income for most (80 per cent) of Australians, an unsurprising figure given that superannuation is a long way from maturity. But projections of the future take-up rate (2050) are only marginally lower (by 5 per cent) than the current rate. However, the balance between full- and part-rate pension recipients will have reversed itself, with two-thirds receiving a part-rate pension and one-third receiving a full pension (Commonwealth of Australia 2002).

The turn of the century also witnessed continued contest over whether the mandatory 9 per cent SG employer contribution will yield an adequate retirement income. The fact that Social Security social insurance contributions in the United States are 12.4 per cent and average 16 per cent in the OECD countries suggests, in concert with Paul Keating's vision of an eventual 15 percent contribution level (including 3 per cent contributed by government and a further 3 per cent by the employee), that the current SG contribution is inadequate. And this is certainly the view of ASFA and the ACTU.

In sum, then, despite scope for still further reform, Australia entered the 21st century with the building blocks fairly firmly set in place of a reasonably coherent retirement income system designed to replace pre-retirement earnings. And yet, it is possible to identify a number of policy initiatives that reflected the beginnings of a return to an era of pronounced sketchiness or fuzziness in this policy domain.

The Re-Emergence of the Slippery Fish

This renewed slipperiness in the retirement income policy fish is reflected in the emergence of a new, fourth pillar to Australia's retirement income system as well as a strengthening and reshaping of its third pillar, changes which largely coincided with the turn of the century.

The Emergence of a Fourth Pillar

The new, fourth pillar is that of 'non-retirement' or, to be more precise, delaying complete retirement.

Public policy initiatives to encourage workers to delay retirement are largely the result of the confluence of concerns about (1) the public costs of population ageing and (2) current, but especially future, labour shortages. Current labour shortages are a result of Australia's long economic boom. In the future, however, population ageing will cause sharp declines in the labour force participation rate and in the number of new labour force entrants. Labour shortages arising from population ageing will start to be felt in a few years time when members of the front end of the baby boom start to retire, an effect that will be compounded by labour shortages should good economic times continue.

One of the initiatives introduced (in 1998) to encourage delayed retirement was the Pension Bonus Scheme. (9) Another initiative designed to encourage delayed retirement was the introduction, in July 2005, of Transition-to-Retirement (TFR) arrangements for those who wished to continue to work part-time rather than fully retire. Under TTR people can access part of their superannuation lump sum once they have reached preservation age without having to leave the workforce provided it is used to purchase an income stream. The emergence of this fourth pillar of Australia's retirement income system is reflective of ...

   ... a loosening of the traditional lifecourse boundaries associated
   with retirement ...--a reconstruction of later life where a new
   series of expectations and behaviours [i.e., continued labour
   force participation rather than increased leisure] are being set
   in place. [This is] an attempt to shape a new ageing citizen, a
   construction that is in tune with changing population trends.
   (Biggs, Phillipson, Money & Leach 2006: 239-240).

There is extensive precedent in Australia and overseas for employing various social policy initiatives to regulate the supply of labour (Borowski 1990). And so, to the last line of the quote should be added: "... as well as changing current and projected labour market circumstances."

A Strengthening and Reshaping of the Third Pillar

Voluntary personal savings (including owner-occupied housing) has generally been seen as the third pillar of Australia's retirement income system. The turn of the century also largely coincided with the beginning of a nascent shift in policy direction relating to this pillar. This can be seen, for example, in the introduction of the spouse rebate in July 1999, the co-contribution scheme in July 2003 and the expansion, from July 2004, in the segment of the population able to make tax-deductible contributions to a superannuation fund. The first two measures were designed to enhance the retirement incomes of people on lower incomes. (10)

Still another manifestation of this nascent policy shift was the extension, in 2004, of tax deductibility (within age-based limits) for contributions made to superannuation by anyone between 18 and 65 years of age. Indeed, even early retirees who wished to top up their superannuation could continue to do so until age 65 and claim a tax deduction for the first $5 000 and then 75 per cent of the rest up to the then prevailing age-based limit.

While all of these provisions seek to strengthen the third pillar of voluntary personal savings, none of them require that the beneficiaries--the recipients of the spouse contribution, the co-contribution or tax deduction--are labour force participants. Thus, this strengthening of the third pillar has involved a weakening of the nexus between work and superannuation as a vehicle for replacing earned income in retirement. Further, all three provisions are premised on financial capacity--of one spouse to contribute to the other's superannuation account or to make a contribution in order to attract either the government's co-contribution or tax deductibility. Thus, while the beneficiaries of the spouse rebate and co-contribution were intended for people with low incomes, all three measures are publicly supported savings opportunities enjoyed most by those who have the financial wherewithal to do so.

A report prepared by the Parliamentary Library (Commonwealth of Australia 2005) illustrates both the erosion in the work-superannuation nexus and the strengthening of public support for savings by the more advantaged. It showed that of the 606 000 people who obtained co-contributions averaging about $540 per person, the partner of a significant proportion had taxable incomes well above the scheme's designated upper income threshold. Further, it is likely that many recipients were not in the workforce. And the same report cites an ACTU submission which points out that the take-up rate of the co-contribution scheme is perhaps less than 10 per cent of those eligible, suggesting either ignorance of the scheme or a lack of discretionary income available to make the contribution required to attract the co-contribution.

But while the emergence of a fourth pillar and the re-shaping of the third pillar suggest policy slipperiness, they almost pale into insignificance relative to the superannuation changes announced in the 2006 Federal Budget.

The 2006 Federal Budget

Most of the changes in Australia's superannuation arrangements announced by the Federal Government in the 9 May 2006 Budget took effect on 1 July 2007. They have been described 'as the largest overhaul of Australia's retirement saving system since the advent of the compulsory superannuation system in the late 80s and early 90s' (Nielson 2007: 4). Their intent was to 'sweep away the current raft of complex tax arrangements, improve retirement incomes and increase incentives to work and save' (Costello 2006). The devil, however, is in the implications of their detail. Indeed, the 2006 Budget changes represent a major reshaping of Australia's retirement income system. While the 'turn of the century changes' outlined above (the emergence of a fourth pillar and the reshaping of the third) represent the beginnings of policy slipperiness, the 2006 changes are of a much higher order. They represent the slipping away of the policy fish--the loss of a reasonably clear sense of the goals and roles of a group of pillars whose presumptive roles had, hitherto, been to contribute to the replacement of pre-retirement earnings.

The focus of the 2006 Budget changes, tagged "Simplified Superannuation" (SS) (or the 'Better Super Reforms') was the introduction of new limits on contributions and the taxation of benefits when paid to the individual.

Contributions

Contributions to superannuation funds can be made on a before-tax basis where they are tax-deductible to the payer (typically an employer). These are known as tax deductible or concessional contributions. Alternatively, they may be made on an after-tax basis where they are not tax-deductible to the payer. The payer here is typically an employee and the contributions are known as non-deductible or non-concessional contributions. Tax-deductible contributions are included in the taxable income of superannuation funds and are taxed at a rate of 15 per cent (the 'entry tax' previously referred to). Personal after-tax superannuation contributions are not tax-deductible and are not included in the income of superannuation funds and, hence, are not subject to the entry tax.

Under the 2006 Budget changes non-deductible (i.e., after tax) contributions made by individuals are limited to $150 000 per year or up to $450 000 in one year as an average over a three year period. (11) Non-deductible contributions in excess of these limits are taxed at a rate of 46.5 per cent. With regard to tax deductible (or concessional) contributions made by an employer on behalf of an employee (and by the employee on a salary sacrifice basis), these are limited to a total of $50 000 per year, or $100 000 per year for those aged 50 and over under special transitional arrangements covering the years 2007-08 through to 2011-12. Tax-deductible contributions in excess of these annul limits are now taxed at 31.5 per cent. With regard to the self-employed, all personal superannuation contributions can be claimed as a tax deduction.

A further transitional arrangement was a provision for non-deductible contributions of up to $1 million made between 10 May 2006 and 30 June 2007. This was intended to allow people who were planning to make a large contribution under the then existing rules to do so. This provision resulted in a surge, conservatively estimated at between $10 billion and $15 billion, in superannuation contributions by high-income earners in the June quarter of 2007, a level of contributions five time greater than the June quarter of 2006 (Uren 2007).

Taxation of Superannuation

Superannuation benefits may comprise tax free and taxable components. The latter are usually made up of tax deductible contributions and earnings on those contributions. The May 2006 Budget foreshadowed the abolition, effective 1 July 2007, of the tax on superannuation benefits (the exit tax), whether the benefits are received in lump sum or pension form, for recipients aged 60 and above. For those who receive their benefits between preservation age and age 59, the first $140 000 (previously $129 751) is tax free while the balance is taxed at 15 per cent. Recipients of a superannuation pension who are between preservation age and age 59 are taxed at their marginal tax rate but with a 15 per cent tax offset.

A Critique of the 2006 Budget Changes

There are a number of major implications of the 2006 Budget changes. These include: (1) the opportunities the changes open up for tax minimization and avoidance, (2) their accentuation of pre-existing inequities, (3) their nurturance of inter-generational inequities, (4) contest about whether they will encourage older workers to remain in the workforce, (5) the blurring of the role of the Age Pension and (6) and their deleterious future effects on the public purse.

Tax Minimization and Avoidance Opportunities

One of the criticisms of the 2006 changes is that they open the door for both tax minimization and tax avoidance--for tax arbitrage and for putting income through a 'tax washing machine.'

The notion of arbitrage refers to 'taking advantage of a price differential between two or more markets' (Fenech 2007). In the context of the SS changes, tax arbitrage arises, for instance, where a worker 60 years of age or over, provided she remains within the new concessional superannuation contribution limits (including the 9 per cent SG), salary sacrifices her entire earnings down to the income tax-free threshold and then receives tax-free superannuation payments under a transition-to-retirement (TTR) arrangement to supplement her income. In this scenario the only tax payable is the 15 per cent superannuation contributions tax. TTR, whereby a person continues to work (and thus accumulate superannuation contributions) while simultaneously drawing a superannuation pension, was originally intended for those who wanted to reduce their working hours but maintain their spending capacity. Since the 2006 budget changes, however, 'it has morphed ... for the over 60s [worker] into a full-blown tax arbitrage opportunity' (Fenech 2007: 40).

Clearly, the SS changes confer a tax-preferred status on those 60 years of age or over. This status potentially lends itself to tax avoidance. An enduring verity of tax planning is 'that as soon as an entity is given some special tax status there is an inevitable interest in using that status by those who do not have [it]' (Haggstrom 2007: 75). Under SS those who do not enjoy this status (those less than 60 years of age) could conceivably use superannuation as a tax washing machine. For instance, an older parent could be given a sum of money by an adult child to contribute to a superannuation fund. This sum and its earnings could then be withdrawn at a later stage completely tax-free and returned to the adult child. Such tax avoidance is obviously 'offensive from a policy perspective' (Haggstrom 2007: 75).

The Accentuation of Inequities Among Low- and High-Income Workers

Another criticism of the SS changes is that they have greatly accentuated the inequity between low-/middle-income and high-income earners arising from the tax treatment of superannuation. While long-recognized and objected to by advocacy groups and policy analysts (e.g., by ACOSS 1988 & 2002, Disney 2007), this inequity, in public discourse, has been largely ignored.

The inequities associated with the tax treatment of superannuation arise in relation to both contributions and superannuation fund earnings. First, the contributions tax, rather than being progressive, is imposed at a constant rate of 15 per cent. Second, if one considers employer superannuation contributions as a form of deferred compensation then, for a high income earner, the contribution tax is much less than the marginal income tax rate that would otherwise have been paid on the employee's earnings. This is also true of voluntary salary sacrifice contributions made on a pre-tax basis by the employee. A similar inequity exists in relation to superannuation fund earnings. The 15 per cent earnings tax is much lower than the income taxes that would be payable by high income earners on other forms of investment income, such as bank interest. Clearly, the superannuation tax regime is biased in favour of higher income earners. As Disney (2007) points out, an equitable system would be one in which a higher public (tax) subsidy is offered to low- and middle-income earners than high-income earners.

The SS changes do not address these inequities. Indeed, they do little to better encourage and support lower- and middle-income workers to save for retirement. Since there is no tax on lump sums up to $140 000, an amount well above current average superannuation account balances, the financial gains flowing from the removal of the exit tax for those aged 60 and over will largely accrue to higher income earners, i.e., those who expect to receive a lump sum in excess of this threshold amount.

The Nurturance of Intergenerational Inequity

A third criticism of the 2006 Budget changes is that they serve to further underscore an 'old' criticism of Australia's mandatory superannuation arrangements, namely, that it unfairly privileges people whose need for savings arises in retirement rather than only or also in earlier years. And this is so even though income needs in retirement may be reduced by using savings for the purchase of assets (especially a family home) or investment in human capital (education and training) earlier in life (Disney 2007). (12)

The acknowledgement of these needs and the role that tax-subsidized savings accounts can play in meeting them is reflected in (then) Labor Opposition Leader (and now Prime Minister) Kevin Rudd's July 2007 proposal (and reiterated in November 2007 during the election campaign) to establish tax-advantaged home deposit savings accounts for first home-buyers. Mr Rudd's plan included provision for salary sacrificing of pre-tax earnings into a home deposit account in which all monies must be held for at least four years. In a similar vein, Howe (2007) proposed a training equivalent to superannuation, i.e., lifelong learning accounts made up of employer, employee and government contributions of one per cent each to finance further education and training throughout one's working life.

The large financial windfall provided to those 60 years of age and over by the SS changes once again calls attention to the unfair privileging of retirement income needs over those arising at earlier life stages.

Incentives for Labour Force Participation

One of the objectives of SS was to increase work incentives. Australia is currently experiencing major labour shortages, a situation that will become even more acute in the future as progressively steeper declines in the number of new labour force entrants broadly coincide with growing momentum in the retirement of the baby boomers. The Government's response has been to increase immigration (currently about 180,000 a year) and bolster the numbers entering Australia on temporary work visas. Another strategy has been to try to lift labour force participation rates, including those of women with children and, as noted above, older workers. (13) Higher participation rates for older workers will also help contain the public costs of supporting them in retirement.

Intuitively, the prospect of tax-free superannuation benefits from age 60 would serve as a strong incentive to remain in the work force, at least until age 60. However, Freebairn (2007) has questioned this. He argues that it is unlikely that the removal of the benefits tax will increase participation or encourage delayed retirement for at least two reasons. First, the abolition of taxes on benefits will affect very few low- and middle-income earners. (Superannuation savings below close to $129,751 were already tax free prior to the 2006 Budget changes.) Second, because the income windfall represented by the removal of the exit tax will help people build up their retirement stock of wealth that much faster, this will induce many people to work less and retire earlier. And the TTR provisions would not, in Freebairn's view, offset the effects of the removal of the benefits tax on incentives to retire earlier.

The impact of SS on the incentive to work thus remains an open question.

The Role of the Age Pension: Towards Fuzziness

The Age Pension means test is comprised of both an income test and an assets test. The pension rate is calculated under each test with the test that yields the lower rate being applied. Most pensioners have payment assessed under the income test.

The 2006 Budget provided, effective 20 September 2007, for a halving of the assets test taper (or benefit reduction rate) from $3 to $1.50 for every $1,000 in assets beyond the level that would otherwise have yielded entitlement to a full-rate Age Pension. The reduction in the taper was projected to benefit around 165 000 current Age Pensioners and add perhaps 50 000 new Age Pensioners to the rolls. The latter figure was expected to grow over time.

Under the new assets test, if total assets (excluding owner-occupied housing) are less than $529 250 for a single person and $839 500 for a couple then a small part-rate pension is still payable. Since access to other benefits is tied to Age Pension eligibility (e.g., cheaper medicines, discounted utilities bills and public transport), even a small part-rate pension can be very valuable to the recipient.

It has not been possible to identify, any source that provides an explicit rationale for this liberalization of the assets test. Its September 2007 introduction can reasonably be construed as a pre-election 'sweetener' designed to influence the vote of senior Australians at the November poll (Borowski, Hudson & McMahon, in press). But be this as it may, this liberalization is completely at odds with the objective of reducing reliance on the Age Pension as a source of retirement income. It is also at odds with the Coalition Government's rhetoric about containing the future cost of the Age Pension as the population ages.

As Negline (2007: 9)) observes in relation to the liberalized assets test,

   Most people think the age pension exists to provide the
   not-so-well off with a [minimally] decent standard of living.
   ... But the age pension pays income and also exists to provide
   important benefits to well-off retirees.

The generous assets test limits arising from the easing of the taper suggest that there is more than a kernel of truth to this observation. It follows that what appeared over the period since the 1980s to be an increasingly residual construction of the role of the Age Pension is now being deconstructed. Indeed, the second Intergenerational Report released in 2007 (Commonwealth of Australia 2007) projects the proportion of the aged not receiving any Age Pension in 40 years time as rising only slightly. The balance, however, between full- and part-rate pensioners is projected to shift with those receiving a full pension falling and those with a part-rate pension increasing significantly. This shift notwithstanding, the Age Pension now appears to be seen as here to stay and to play an integral role in providing a retirement income to most Australians well into the future. The seemingly new role of the Age Pension is reflected in Rothman's (2007) estimates of the future adequacy of Australian retirement incomes, estimates which presuppose a continuing, albeit not clearly specified, role for the Age Pension. Notions that a progressively maturing superannuation system would substantially obviate the need for the Age Pension in providing a retirement income to Australia's workers (Manning 2005: 31) appear to have gone out the window! What its precise role should now be, however, is unclear.

Future Effects on the Public Purse

A final criticism of the 2006 Budget changes is that they will have long-term adverse effects on the public purse.

According to ACOSS (2006), superannuation tax concessions and the co-contribution scheme cost the government $17.3 billion in 2006-07 while the Age Pension cost a further $22.1 billion. It has been estimated that the 2006 Budget changes will add a further $7.2 billion over four years (Steketee 2007). The revenue loss associated with superannuation tax concessions is projected to continue to grow (mostly due to the higher cost of tax concessions on superannuation fund earnings) and outstrip the growing cost of the Age Pension. The rising costs of these tax concessions is clearly at odds with what hitherto had been the goal of reducing the public subvention of superannuation. In addition, this huge revenue loss 'will make it more difficult for future Governments to provide the necessary health care and other services needed by future retirees' (ACOSS 2006: 9) and place a heavier-than-otherwise support burden on future workers.

While retirement income policy has largely enjoyed bipartisan support over the course of the last few decades, doubts about the long-term sustainability of tax-free superannuation benefits for those 60 years of age and over led some observers to suggest that a Labor victory at the federal election on 24 November would result in removal of this generous provision. Given the popularity of tax-free superannuation among older Australians, the Opposition was silent on this matter during the 2007 election campaign, a silence that served to feed speculation of a rollback by an incoming Labor government (Flint 2007, Murray 2007). However, in March 2008, the new Labor Superannuation and Corporate Governance Minister, Nick Sherry, announced that the pervious government's SS changes would not be touched (Negline 2008: 5).

Where to From Here?

The modern history of Australian retirement income policy is one that can be characterized as involving efforts by successive federal governments to grasp a previously slippery policy fish, succeeding in placing a reasonably firm hand on it--bringing clarity to the goals and roles of its several pillars--only to then later allow that fish to become slippery again and start to slip away.

Today Australia's superannuation schemes are no longer purely a tax-advantaged savings mechanism designed to contribute to the replacement of workers' pre-retirement earnings. They have morphed into a more generalized savings mechanism that can also be utilized by people who are outside the workforce. While many countries in the world have established tax-advantaged savings schemes for a variety of purposes, Australia's superannuation arrangements privilege the income needs of older Australians including those whose need for income no longer necessarily arise as a result of withdrawal from the workforce upon retirement.

And it is a highly unfair mechanism. The inequitable tax treatment of superannuation requires attention to overcome its inherent bias in favour of high income earners. ACOSS (2006), for example, has recommended that all contributions be made from after-tax income thus levelling the playing field between contributions from high-, middle- and low-income earners. ACOSS has further suggested that the tax concessions be replaced with a simple tax offset paid annually to the superannuation fund which would provide most assistance to those whose need for public support is greatest and to encourage voluntary saving.

Superannuation arrangements also involve enormous costs to the public purse today and well into the future and will thus impose a sizeable burden on tomorrow's workers. They also impact negatively on the public purse through lending themselves to tax minimisation and avoidance.

While incentives to increase superannuation contributions for low-income people have been introduced, they also advantage the more advantaged because participation in these programs requires a financial capacity which may not, in fact, exist among the target population.

And perhaps the greatest conundrum that arises from the spate of changes in Australia's retirement income arrangements concerns the precise role of the Age Pension. Rather than substantially withering away and being available to the neediest as Australia's superannuation system matured (and thereby also reducing the public cost of population ageing), it is now seen as integral part of retirement income well into the future. While there is ample evidence indicating that large numbers of older Australians remain in poverty (Senate Community Affairs References Committee 2004, Headey, Warren, & Harding 2006, ACOSS. 2007, ABS 2007b) and that cost of living pressures on older Australians are growing (Commonwealth of Australia 2007b, National Seniors Association 2007), the role of the public pillar in Australia's retirement income system has become quite unclear. Eligibility liberalizations (most recently of the asset test taper) has meant that the Age Pension, albeit increasingly part-rate, will be widely available in the future.

Australia's retirement income system has always been a complex one. The whys and wherefores of its constituent elements--its pillars--and their interaction have been hard to understand. Today, it is a system not exclusively geared towards contributing to the replacement of pre-retirement earnings of workers, is highly inequitable, enormously costly to the public purse and lacks coherence. What is needed is for a new group of zealous reformers, of the ilk of those who sought to forge a coherent system beginning in the 1980s, to step into the fray. Their charge will be to once again construct a vision of a future system and to carefully articulate the role of its various pillars in realizing the envisioned ends.

References

ABS. (1998) Employment Benefits Australia 1983-1992 (Cat. no. 6334.0) Australian Bureau of Statistics, Canberra.

ABS. (2005) (18 March) Year Book Australia: Labour Earnings and Benefits, viewed 30 November 2005, <http://www.abs.g0v.au/ausstats/abs@nsf/0/6C28.AAF19A87C96CA256F7200832F>

ABS. (2007a.) Employee Earnings, Benefits, and Trade Union Membership (Cat no 6310.0) April, Canberra, Australian Bureau of Statistics.

ABS. (2007b) 'Low income low wealth households.' In Australian Social Trends 2007 (Catalogue No. 4102.0) Canberra, Australian Bureau of Statistics.

ABS. (2007c) Managed Funds, Australia, Dec 2007 (Catalogue no. 5655.0) Canberra, Australian Bureau of Statistics.

Access Economics Pty Limited for AMP. (2007) The AMP Superannuation Adequacy Index Report, Sydney, AMP Financial Services.

ACOSS. (1988). Reform of Superannuation Tax Concessions, Sydney, Australian Council of Social Services.

ACOSS. (2002). Fairness and Flexibility- Making Superannuation Work for Low-and Middle-Income Earners, Sydney, Australian Council of Social Services.

ACOSS. (2006) ACOSS Submission to Australian Treasury--Proposals to Simplify Superannuation, Sydney, Australian Council of Social Services, August.

ACOSS. (2007) A Fair Go for All Australians Strawberry Hills, NSW, Australian Council of Social Services

APRA. (2007) Celebrating 10 Years of Superannuation Data Collection 1996-2006 APRA Insight Issue 2, Canberra, Australian Prudential Regulation Authority.

ASFA. (2004) Taxes on super (Fact Sheet no. 4) Sydney Association of Superannuation Funds of Australia.

ASFA. (2007) Superannuation Statistics--October 2007 Association of Superannuation Funds of Australia, Sydney.

Barrymore, K. (2007) 'Hardship draining savings' Herald Sun, Monday 23 July, p. 27.

Biggs, S., Phillipson, C., Money, A. & Leach, R. 2006. The Age Shift: Observations on Social Policy, Ageism and eh Dynamics of the Adult Lifecourse Journal of Social Work Practice 20(3): 239-2250.

Borowski, A. (1990) 'The work-leisure choice: The causes of early retirement', Australian Social Work, 43(2) (June), 27-36.

Borowski, A., Schulz, J. H. & Whiteford, P. 1987. Providing and Adequate Retirement Income: What Role Occupational Superannuation? Australian Journal on Ageing 6(1), 3-13.

Borowski, A., Hudson R. & McCormack, J. (In press) Elder Participation and Senior Power in Australian Electoral Politics. Journal of Ageing and Social Policy, 21(1).

Business Council of Australia. (2007) Employing Our Potential: Ensuring Prosperity Through Participation. Workforce Participation Roundtable Discussion Paper. May

Clare, R. (2007) Are Retirement Savings on Track? Sydney, The Association of Superannuation Funds of Australia.

Commonwealth of Australia. (1992)Safeguarding Super, Canberra, Senate Select Committee o Superannuation, June.

Commonwealth of Australia. (2002) Superannuation and Standards of Living in Retirement, Canberra, Senate Select Committee o Superannuation, December.

Commonwealth of Australia. (2003) Planning for Retirement, Canberra, Senate Select Committee o Superannuation, July.

Commonwealth of Australia. (2005) Super co-contribution: Performance to date, (Research Note no. 16) Canberra, Parliamentary Library, Department of Parliamentary Services, Parliament of Australia.

Commonwealth of Australia. (2007a) Intergenerational Report 2007, Canberra, Department of Treasury.

Commonwealth of Australia. (2007b) Official Committee Hansard, Senate Standing Committee on Community Affairs, Reference: Cost of living Pressures on Older Australians, Canberra, Thursday 20 September.

Costello, Peter The Hon ME (2006) 'Simplified superannuation--Final decisions' Media Release No. 093. Canberra, Commonwealth of Australia, 5 September.

Department of Social Security. (1984) Government Support for Retirement Incomes in Australia (Research Paper no. 24), Policy Review Branch, Development Division, Canberra.

Disney, J. (2007) 'Superannuation and lifelong saving'. Paper presented to the From Welfare to Social Investment: Reimagining Social Policy for the Life Course Conference, Centre for Public Policy, Melbourne 21-22 February.

Fenech. A. (2007) 'Super way to pay just 15pc tax', The Weekend Australian, 9-10 June, 40.

Flint, D. (2007) "Rudd will tax your super, where's the outcry? Crikey.com.au. Downloaded 2 November.

Forward, R. (1983) Income in Old Age: Is National Superannuation the Answer? (Current Issues Brief No. 4), Department of the Parliamentary Library, Parliament of Australia, Canberra.

Freebairn, J. (2007) Some Policy Issues in Providing Retirement Incomes. Melbourne Institute Working Paper Series, Working Paper No. 6/07, Melbourne, Melbourne Institute of Applied Economic and Social Research, The University of Melbourne.

Haggstrom, P. (2007) 'Super changes the intergenerational contract', Australian Financial Review 8 June, 75.

Headey, B., Warren, D & Harding, G. (2006) Families, Incomes and Jobs: A Statistical Report of the HILDA Survey Melbourne, Institute of Applied Economic and Social Research.

Hughes, D. (2005) Keating urges plug for $600bn super hole', The Age, 10 March, 3.

Howe, A. (ed.) (1981) Towards an Older Australia, The University of Queensland Press, St Lucia.

Howe, B. (2007) Weighing Up Australian Value, Sydney, UNSW Press.

IFSA. (2006) Retirement Incomes & Long-term Savings Policy Options, Sydney, Investment and Financial Services Association.

Kingson, E.. & Williamson, J. B. (2001) Economic Security Policies. In R. H. Binstock & L. K. George (eds) Handbook of Aging and the Social Sciences, San Diego, Academic Press, 5th edn, pp. 369-86.

Manning, I. (2005) 'Retirement incomes: National superannuation, social insurance or something else?' Transitions in Australian Labour Markets-Initial Perspectives (CEDA Information Paper 82) Melbourne, pp. 29-34.

Murray, P. (2007) 'Can Rudd be trusted with your super cash?' The West Australian 26 October. Downloaded from thewest.com.au on 4 November.

Myles, J. & Pierson. P. (200) 'The Comparative political economy of pension reform'. In P. Pierson (ed.) The New Politics of the Welfare State, New York, Oxford University Press, pp. 305-33.

National Seniors Association. (2007) Submission to Parliament of Australia, Senate Community Affairs Committee, Inquiry into the Cost of Living Pressures on Older Australians, Deakin, ACT, National Seniors Association.

Negline, T. (2007) Investment balancing act. Wealth, The Australian, Wednesday 24 October, p. 9.

Negline, T, 2008. Reduce the taxman's cut. Wealth, The Australian, Wednesday 23 April, p. 5.

Nielson, L. (2007) Superannuation and Taxation 2007-08 Research Paper no. 2, 2007-08, Canberra, Parliamentary Library, Department of Parliamentary Services, Parliament of Australia.

Olsberg, D. (1997) Ageing and Money: Australia's Retirement Revolution Sydney, Allen and Unwin.

Patten, S. (2007) 'Super falls for third time', Australian Financial Review, Friday 24 August, 73.

Rein, M. (1999) 'Dynamics in the comparative analysis of social policy reform'. Keynote address presented at the 15th Yokohama 21st Century Forum: Towards an Active Society--Perspectives for Longer Employment and Productive Aging, held in Yokohama, Japan 21-22 January.

Rothman, G. (2007) 'The adequacy of Australian retirement incomes--New estimates incorporating the Better Super reforms'. Paper presented at the Fifteenth Colloquium of Superannuation Researchers held at the University of Sydney, 19-20 July.

Schulz J. H. (2005) Old-Age Income Security: Australia Tries a Different Way Washington DC, AARP.

Schulz, J. H. & Borowski, A. (2006) Economic Security in Retirement: Reshaping the Public-Private Mix. In R. H. Binstock & L. K. George (eds) Handbook of Aging and the Social Sciences, San Diego, Academic Press, 6th edn, pp. 360-79.

Senate Community Affairs References Committee. (2004) A Hand Up and Not a Hand Out: Renewing the Fight Against Poverty. Report on Poverty and Financial Hardship. Canberra, Parliament of Australia.

Shanahan, D. & Karvelas, P. (2007) 'Nursing to lead PM's fightback', The Australian, 14 September.

Stanford, J. D. ((2003). 'Is superannuation safe? The background and the issues', The Australian Economic Review, 36(1), 79-88.

Steketee, M. (2007) 'Younger taxpayers caught by a super stab in the dark' The Australian 5 April.

Uren, D. (2007) Wealth put $15bn in super scramble, The Australian, Friday 31 August, p. 3.

Warren, N. (2004) Tax: Facts, Fiction and Reform (Research Study no. 41) Sydney, Australian Tax Research Foundation.

Whiteford, P. & Bond, K (2000) Trends in the Income and Living Standards of Older People in Australia (Policy Research Paper no. 4) Department of Family and Community Services, Canberra.

World Bank. (1994) Averting the Old Age Crisis: Policies to Protect the Old and Promote Growth New York: Oxford University Press.

(1) Superannuation had played such a minor role in providing a retirement income that Dixon and Foster (1982) wondered why such schemes existed at all! They were also at a loss to explain how policy toward it had developed.

(2) The preservation age is the age at which superannuation benefits can be accessed provided that the fund member has permanently retired.

(3) See Olsberg (1997) and Schulz (2005) for a more detailed analysis of why superannuation was mandated.

(4) Because of a 15 per cent contributions tax, the 9 per cent contribution is effectively a net contribution of 7.65 per cent.

(5) This includes employees under 18 years who work less than 30 hours weekly and employees who earn less than $450 per month from a single employer.

(6) A fuller treatment of these and other risks appears in Schulz and Borowski (2006).

(7) The adequacy of the retirement benefits that superannuation will yield in the future is contested. This contest derives from differences in what is included in both the numerator and denominator of the income replacement rate that is used for assessing adequacy. Thus, research completed by actuaries Rice Walker for IFSA estimated in 2006 that the national shortfall in income required to fund a comfortable retirement was $450 billion (IFSA, 2006). This finding is underscored by those of another study which reported that the average superannuation payout in the future at age 60 for average income earners 35-44 years of age and receiving the 9 per cent SG contribution only would be just $183,000 for males and $93,000 for females (Clare 2007). In yet another study, Access Economics (2007) reported that about a quarter of retirees will have enough superannuation to enjoy a self-funded retirement without the Age Pension. In combination with other financial assets additional to superannuation, however, two-thirds of the workforce can expect to retire with an income equivalent to 2/3 (65 per cent) of their incomes in their final year of work. And finally, Rothman (2007), employing estimates that take account of the 2006 Budget changes, reports that the typical retiree now receives 52 per cent of pre-retirement income in retirement. This will rise to 60 per cent in 2010 and to 70 per cent by 2015.

The first two studies exclusively focus on the role of superannuation in replacing pre-retirement earnings while the latter two base their adequacy estimates on pre-retirement income that includes both earned and unearned income and post-retirement incomes that include superannuation and, as integral part of most retirement incomes, both the Age Pension and other sources of unearned income.

(8) In addition to preservation and portability requirements, the 1987 OSSA legislation provided, for example, for the vesting of benefits and greater member involvement in the control of superannuation funds. Subsequent to the introduction of SG contributions in 1992, OSSA was replaced, from 1 July 1994, by the Superannuation Industry (Supervision) Act (SISA). SISA, among other things, improved disclosure and regulatory reporting requirements, set out the duties and responsibilities of fund trustees and enlarged auditors' and actuaries' roles. And in July 1998, the Australian Prudential Regulatory Authority came into being to supervise the banking, insurance and superannuation sectors (in the place of the discontinued Insurance and Superannuation Commission) while the newly-created Australian Securities and Investment Commission (ASIC) assumed responsibility for disclosure and other consumer protection issues. Changes in regulatory legislation remain ongoing in response both to changes in retirement income policy and the identification of challenges to the safety of superannuation savings arising, for instance, from the corporate collapses and the sorts of questionable practices of financial institutions that have come to the fore during the current sub-prime mortgage market crisis.

(9) This scheme awards those who are Age Pension-eligible a maximum tax-free lump sum of up to $32,000 for a single person and $27,000 for each member of a couple who defer claiming the Age Pension and continue working for up to five years beyond pension age. About 57,000 people have received an average cash bonus of about $12,000 since 1998 and a further 66,000 are currently using the scheme. In September 2007, Family and Community Services Minister Mal Borough indicated that he would push the Government to provide access to the Pension Bonus Scheme to retirees who decide to return to work (Shanahan & Karvelas 2007).

(10) Under the spouse rebate, a contributing spouse can receive a tax rebate of 18 per cent for contributions to a superannuation account made on behalf of a spouse (up to a maximum of $3 000 per annum) whose income is $10 800 or less per annum. The maximum rebate of $540 phases out on a dollar-for-dollar basis and ceases when the spouse's income reaches $13 800.

Under the non-means-tested co-contribution scheme, the government contributes $1.50, up to a maximum of $1 500, for every dollar in after-tax contributions made to a superannuation fund by people with an income of less than about $29 000 (in 2007). The rate of co-contribution progressively decreases until it eventually ceases when the individual's income reaches close to $59 000.

(11) Since it is possible to contribute $450 000 at the beginning of one three-year period and again at the beginning of the next, in reality, it is possible to contribute $900 000 within a three-year period.

(12) The need for such savings is already recognized in provision for the release, subject to APRA approval, of retirement savings on hardship and compassionate grounds. In 2006, $135 million in toto was released to 16,500 people (Barrymore, 2007).

(13) For a more detailed discussion of strategies to increase labour force participation rates see Business Council of Australia (2007).