TRR #5 (2026): The Retirement Gap Starts Long Before Retirement

Welcome to TRR #5 for 2026. This year’s International Women’s Day theme Balance the Scales is a fitting frame for retirement income, where the scales often begin tilting well before retirement arrives.

Unequal retirement outcomes do not begin at retirement. They arrive there, built up over decades of unequal pay, interrupted careers, unbalanced carer duties, household disruption and eroding housing security. For trustees designing retirement income solutions, that distinction matters enormously.

The question is no longer whether these differences matter. It is whether retirement strategies are adapting fast enough to reflect them.

This year’s International Women’s Day theme, Balance the Scales, is a timely reminder that financial inequality in later life is usually the cumulative result of disadvantage much earlier in life. In retirement, that shows up not only in super balances, but in the circumstances in which people reach retirement and the options available to them once they get there.

That was one of the clearer findings of the 2020 Retirement Income Review. Its assessment of gender and partnered status made the central point plainly: the retirement savings gap between men and women largely reflects the economic disadvantages women face during working life. The Review also noted that the working-life earnings gap is the main driver of the super gender gap at retirement, and that women are more likely to enter retirement single and to spend longer in it.

That framing remains highly relevant in 2026. The newly released WGEA Employer Gender Pay Gap report shows that the average total remuneration gap was still 11.2 per cent in 2024–25, even after a modest improvement from the prior year. In other words, the structural gap may be narrowing, but it remains large enough to shape retirement outcome differences between the genders.

As the advocacy group ‘Women in Super’ rightly notes, superannuation is built on earnings across a lifetime. That makes pay equity a long-run retirement policy issue, not simply a workplace one.

The significance of the gender pay gap is not confined to income while working. Lower earnings do not just reduce compulsory super contributions. Over time, they also slow wealth accumulation, weaken buffers and leave less room to absorb financial disruption.

The 2024 Women’s Budget Statement makes this point clearly.

That matters because weaker wealth accumulation tends to amplify the impact of later-life shocks. By the time retirement approaches, the issue is no longer simply that balances are lower. It is that lower balances may sit alongside more fragile housing outcomes, fewer household economies of scale and less flexibility to respond to disruption.

Recent work by ASFA adds an important dimension to the gendered retirement discussion, particularly in relation to separation, divorce and housing. Its newly released research paper, The Super Split: The reality for Australian women and superannuation, brings into focus the extent to which retirement outcomes are shaped not just by earnings history, but also by what happens to households and housing later in life.

The paper notes that 49,241 divorces occurred in Australia in 2022, with the median age at separation for women being 39.8 and the median age at divorce 43.7. For men, the respective ages were 42.6 and 46.7. These are not peripheral life events. They occur during the very years in which super balances, housing positions and retirement trajectories are being consolidated.

The Retirement Income Review had already surfaced this dynamic. In 2016, 19 per cent of women and 15 per cent of men aged 60 to 64 were divorced and single, up materially from 1991. That matters because single retirees generally have lower rates of home ownership, lower wealth and less capacity to absorb financial stress than couple households.

ASFA’s use of HILDA data is particularly helpful because it takes the discussion beyond balances and into housing tenure. For divorced women and men approaching or entering retirement, the share owning a home outright is much lower than traditional retirement assumptions would suggest.

Housing tenure of divorced women and men aged 60 to 69

For single-income households with a mortgage, the median balance outstanding is also significant. ASFA notes that for men aged 50 to 59 the median mortgage balance was $270,000, dropping to $159,000 for those aged 60 to 69. For women, the pattern is striking: $187,500 for those aged 50 to 59, rising to $230,000 for those aged 60 to 69.

Taken together, these findings suggest that for many older Australians, especially those who are single after separation or divorce, retirement risk is increasingly shaped by housing position as much as by superannuation balance.

Gendered retirement outcomes cannot be treated within the narrow lens of super balance. The evidence points to a broader reality: retirement adequacy is increasingly shaped by the interaction between income, partner status, housing tenure and household resilience.

For trustees, that matters because these are not peripheral variables. They affect how members experience retirement risk, how they might respond to guidance services, and what kinds of pathways are likely to be relevant or effective.

This is not just an equity issue, it’s a design issue. And it suggests that age-and-balance segmentation on its own will often be too blunt to capture the members most exposed to retirement risk.

The ASFA paper was prepared, in part, to argue against using superannuation before a condition of release to help separated or divorced Australians buy a home. Whatever one’s view on that policy question, the paper does something else that is highly valuable: it acknowledges the reality of today’s housing landscape and the extent to which it now sits at odds with established retirement norms.

For many years, retirement adequacy frameworks implicitly leaned on the assumption of debt-free home ownership at retirement. That assumption is becoming less defensible, particularly for lower-and-middle wealth singles, and for those whose working lives and household lives have been more disrupted.

ASFA itself notes that in 2019–20 only around 70 to 80 per cent of households over 65 owned their property outright, with mortgaged owners and renters making up the balance. Overlay that with the paper’s findings on divorced older Australians and the picture becomes sharper still: among divorced women and men aged 60 to 69, outright ownership is far from the norm.

In 2023, we described the disconnect between retirement assumptions and housing reality as the ‘Elephant in the Super Retirement Room’. It is encouraging to see that issue now receiving more direct attention. What was once treated as an exception increasingly looks like a structural feature of later-life retirement risk.

Given this improved understanding of housing tenure for older Australians, particularly older singles who are separated or divorced and not re-partnered, what follows for trustees?

Partner status and housing tenure should increasingly inform cohort analysis, not just age and balance. If trustees are serious about designing retirement income solutions that reflect actual member circumstances, these variables matter.

APRA and ASIC have consistently encouraged trustees to better understand the circumstances of their members leading up to, and in, retirement. Findings such as these reinforce why closing those data gaps matters. The member experience at retirement cannot be understood well if major drivers of retirement adequacy remain invisible.

The practical challenge, which we readily acknowledge, is to build systems and processes that can better reflect partner status and housing position while staying on the right side of the general versus personal advice boundary. That is no trivial undertaking.

But it is precisely the kind of capability that forward-thinking trustees will need if they want to deliver more integrated retirement income solutions to diverse memberships over time.

Our view (outlined in the previous newsletter) is that Treasury’s best practice principle guidance of a minimum of three cohorts is a floor, not a ceiling. Funds with sociographically diverse memberships will likely need more.

For trustees, the challenge is not just in recognising these differences, but in building the systems, segmentation and member journeys needed to respond to them. That is where better data strategy, sharper cohort design and stronger guidance pathways can support better retirement outcomes.

If improving member segmentation or retirement cohort design is on your agenda, we’d be glad to discuss what sharper data strategies and member journeys could look like for your fund.