Based on the redacted version released recently under Freedom of Information( FOI), DSS’s incoming government brief seems overall to be a valuable document, particularly the volume titled, ‘Strategic Considerations’, which canvasses longer-term challenges for the portfolio.

But its subsection on Social Security Reform is disappointingly narrow, omitting the biggest challenge for our retirement incomes system: how to help retirees convert their superannuation savings into secure regular incomes integrated with any age pension entitlement.

Instead, the brief merely suggests that “low and middle income taxpayers are subsidising the retirement incomes of seniors with significant wealth in addition to their homes”, highlighting the means test cut-out points where eligibility for any age pension stops (just under $100.000 a year in income for a couple or assets of just under $1.05 million in addition to their homes).

Not mentioned is that the cost of the age pension (and related payments for those of age pension age) as a percentage of GDP is in fact falling, as is the percentage of retirees eligible. The 2023 Intergenerational Report projects the costs to fall from 2.3% of GDP in 2022-23 to 2.0% in 40 years’ time (the 2021 IGR reported the cost in 2020-21 was 2.7%). The proportion of people of age pension age receiving the pension was already under 70% in 2022-23 (from the 76% in 2020-21 according to the 2021 IGR) and is projected to fall to about 55% (the Henry Report had coverage of about 80% in 2010 and suggested this would continue). Over 40% currently receive the maximum rate of pension, but this is projected to fall by half, with most of the 55% eligible for some pension only eligible for a part-rate.

The cost is already well below the OECD average and the projected decline contrasts with the OECD projected growth on average of 1.4% of GDP over the next 30 years.

There is a case for including the value of the home (above some high threshold) in the assets test and, as my ANU colleague Ben Phillips suggests, increasing the deeming rate of income from assessable assets to a more realistic level under the income test. But it is wrong to suggest Australia has a big problem of too many rich people getting the pension.

The means test cut-out points cited by DSS are, of course, a mathematical outcome of the maximum rate of pension and the rate to which it is reduced as income and assets increase (above the respective thresholds). Is DSS suggesting higher rates of withdrawal, or lower thresholds, or lower rates of pension?

Phillips suggests lowering the assets test thresholds which were substantially increased in 2017. I am willing to accept that, but only if the taper above those thresholds, which was doubled in 2017, was also reduced. The current taper reduces the pension by $3 a fortnight for every $1000 of assets above the thresholds, equivalent to 7.6% of the extra assets, more than retirees can generally expect to earn from those savings (and more than they could gain from purchasing an annuity). That is, their total retirement income may go backwards the more assets they have.

And remember, superannuation assets are mostly built upon compulsory contributions. To force people to reduce disposable incomes while working and then have the forced savings reduce disposable incomes in retirement is hardly fair or sensible.

The problem is that DSS seems to be looking at the system only through the narrow lens of the safety net it provides. But the age pension is only one of the pillars of our overall retirement income system. It is essential that DSS and Treasury (and external commentators) look at how the different pillars, particularly the age pension and superannuation, operate together to ensure not only poverty alleviation but also the maintenance of living standards at and through retirement for all Australians.

More fundamental reform of the means test is needed to get coherence across these two pillars, along with measures to assist retirees to convert their accumulated superannuation savings into secure incomes that meet their requirements and preferences for the rest of their lives.

The Grattan Institute recommended earlier this year:

• firmer requirements for funds to advise members approaching retirement about the income their savings could deliver rather than just the accumulated amount,
• the government itself offer lifetime annuities for sale and suggest retirees allocate a proportion of their super to such annuities, and
• the government also establish a free advisory service to those approaching retirement.

These recommendations are eminently sensible and should be acted upon quickly, but they are not sufficient.

What is also required is an age pension means test that encourages the more than 50% of retirees likely to remain eligible for some pension to easily convert their superannuation savings into income products that complement their pension entitlement, maximising the combined income and ensuring they can maintain their living standards for the rest of their lives.

The current means test, with separate income and assets tests, is extremely complicated, and the two components are not consistent with each other. A merged means test would be far better under which assessable assets would be converted into the income they could reasonably generate over retirement, and adding that to any other income with the total subject to a single income test. This could be designed with an appropriate taper ensuring extra savings don’t take people backwards and encouraging retirees to convert some of their savings into lifetime annuities paid regularly along with their full or part age pension, providing security and convenience as well as the living standard they want.

Until we ensure retirees are able to confidently convert their super savings into secure retirement incomes which complement any age pension entitlement, our unique retirement income system is not “world leading” as its adherents claim, but “world lagging”.

DSS needs to take a broader view and help the government to complete the reforms Paul Keating began 30 years ago.

First published in The Mandarin, 13 August 2025

The views expressed in this article may or may not reflect those of Pearls and Irritations.