The federal government's superannuation tax changes are mostly pretty modest. They steal much of Labor's policy while making sensible improvements and tying down loose ends.
Yet they are by no means comprehensive reforms to the retirement-income system. Nor has Labor presented such a package. This is surprising, because such a package could be very attractive to most people while further improving long-term financial sustainability.
Central to such a package is to include key attributes of the much-maligned and now-closed public sector benefits-promise schemes: the provision of adequate, regular and secure incomes for life, without the burden of self-management. Some people like playing with investments, most don't, and most who do don't achieve an optimal retirement income anyway. Our top priority now should be to build on the strengths of our fully funded defined-contributions schemes to deliver benefits in the pensions phase that look much more like the ones the old public sector schemes provided (and most overseas social-security schemes still provide). That would give us the best of both worlds.
The key for a coherent and comprehensive retirement-income policy is to focus on the central objective – to provide an adequate income through all the years of retirement for all Australians – and to relate this to each of the three pillars of our system: the age pension, mandated super and voluntary super. The government's commitment to legislate an objective for super is a step in the right direction but does not go far enough as it does not encompass the whole retirement income system.
"Adequacy" for the age pension means a modest, guaranteed, minimum income relative to community norms. The pension achieves this for those above age-pension age who own their homes, but not for those in private rental accommodation or those unable to work to age-pension age. Both political parties should be ashamed of their record here because they have known of the inadequacy of rental assistance and the Newstart Allowance for a long time.
"Adequacy" for most people means more than the age pension. The superannuation system promotes thrift, facilitating the spreading of lifetime incomes to supplement or replace the pension to maintain pre-retirement living standards. Using a benchmark of about 75 per cent net income replacement, actuarial studies suggest the legislated increase in mandated contributions to 12 per cent will achieve such adequacy in future (from age 67) for those whose pre-retirement incomes are around or below median earnings. Those whose incomes are higher than this, who won't be eligible for any age pension, will need to contribute more, as will those planning on earlier retirement or transition to retirement. For these people, the public sector norm of about 15 per cent contributions is in the right ballpark. All this assumes continuing employment; extra contributions may be needed to make up for periods of limited work, particularly for women.
Central to this result is the design of both the superannuation tax regime and the pension means test.
Superannuation tax should do no more and no less than facilitate the spreading of lifetime earnings. If it does this, there should be no complaints about excessive or unfair tax expenditures. The orthodox approach is to exempt contributions and fund earnings, and to tax all benefits in full (which is, in effect, what used to happen with the old public service benefits-promise schemes). The personal income tax scale then ensures that people on high incomes, in retirement and over their lifetimes, pay proportionally more tax and those on low incomes pay much less tax.
While our present tax arrangements – taxing contributions and earnings, and exempting benefits – are unorthodox, the government's proposed changes represent a useful step towards achieving the same overall outcome in terms of lifetime tax and net retirement incomes. Exempting contributions by low-income earners from tax but applying a higher tax on contributions by high-income earners achieves the degree of progressivity we expect from the tax system; indeed, the government could have gone further. The relief for low-income earners will also go some way towards addressing gender equity in the system.
Most of the other measures address the big weakness from not taxing benefits: the risk of subsidising wealth accumulation rather than just facilitating the spreading of lifetime earnings. We wouldn't need caps if we fully taxed benefits.
The new $1.6 million cap, beyond which fund earnings will be taxed in the retirement phase, is hardly draconian. A 15 per cent tax on earnings from additional savings is still less than the tax on returns from most other savings.
The government is also proposing changes to the tax treatment of those receiving defined-benefit pensions (mostly retired public servants and military personnel, and those still in closed schemes), ostensibly to achieve a similar outcome. If the benchmark is the orthodox approach of taxing benefits and exempting contributions and earnings, the pre-Costello regime of taxing defined-benefit pensions in full was exactly right, and the 10 per cent tax rebate that Peter Costello introduced was entirely unjustified. So removing the rebate from defined-benefit pensions above $100,000 is very modest. What remains unclear at this point is the treatment of those defined-benefit pensioners who also have accumulated defined-contribution savings. Presumably, the $1.6 million cap (above which fund earnings are to be taxed) will be reduced through some formula about the value of the defined-benefit pension (e.g. reducing the cap to zero for those whose pensions exceed $100,000 and possibly halving the cap if the defined-benefit pension is $50,000). All a bit messy, but entirely justified.
The other caps the government has proposed are less defensible, but at least the government didn't adopt the Grattan Institute's extreme suggestions. The new $25,000 cap on annual contributions is under the 15 per cent needed for an adequate retirement income for anyone earning more than $160,000 a year, and will constrain the ability of others to make up for periods of limited employment and contributions. How this will affect public servants on defined-benefit schemes with notional contributions above the cap is not yet clear.
Not justified is the proposed $500,000 lifetime limit on non-concessional contributions. Our mandated system works well for employees but not for the self-employed, and a more generous approach to facilitate the spreading of lifetime earnings for those outside the mandated system should be considered.
For those on no more than median earnings, retirement-income adequacy depends on receiving some age pension. The basic design of the means test is appropriate, concentrating assistance on those most in need and containing costs. It's also entirely reasonable to expect retirees to draw down their super assets as well as use the interest those assets earn. But the assets-test taper to come into force next year goes too far and will lead to some people redirecting superannuation savings into housing and/or limiting contributions in later years of their working lives. It wouldn't be hard to redesign the test and still achieve the budgetary savings involved, with a more modest taper and tighter thresholds.
Secure, regular, straightforward retirement incomes
More appropriate use of super savings after retirement is the area where our retirement-income system needs the most attention.
Economic theory tells us that pooling funds leads to more efficient management of risks, such as how long we live and what health and aged-care services we'll need. Yet we now generally don't pool super savings (those in defined-benefit schemes did, and do, of course). The result, as the Murray report revealed, is that people fear running out of money, limit the draw-down of savings and unintentionally leave tax-assisted savings in their estates.
Much firmer policy in this area is needed from both parties. Instead of expressing superannuation savings in terms of accumulated capital, funds should be required to provide regular advice on the income stream those savings could deliver at and through retirement, and the level of further contributions likely to be needed to achieve specific target income levels.
Funds should not only be encouraged to offer retirement-income products that include longevity insurance but should be mandated to make such an offer, as the Murray report recommended. This would ensure that members of all funds can access such products, while also promoting efficiency by encouraging smaller funds to place relevant members' savings into larger funds' pools. The plan should be for these products to become the default options for most retirees, who don't want to be burdened with the complexities surrounding investment strategies, risk management, tax law, means-test strategies and so on.
Our defined-contribution system has many advantages, including financial sustainability and intergenerational equity, but we sorely need it to deliver retirement incomes that are adequate, secure, regular and straightforward.