The retirement execution problem

Welcome to TRR #15 for 2026. This edition examines a stubborn feature of retirement policy: the industry has no shortage of thinking into improving retirement outcomes, but execution is seriously lagging the insights.

A new Monash Centre for Financial Studies study shows how quickly modest balances and poor sequencing can exhaust account-based pensions. The Financial Services Council makes the case that digital advice is now constrained less by technology than by governance confidence. And Margaret Cole’s final remarks as APRA Deputy Chair provide a scorecard on the Retirement Income Covenant: progress, but not at the pace older members need.

Three different vantage points, one recurring theme: insight is abundant in the retirement space; execution remains the harder, and more consequential, work.

Using a stochastic approach, a recent paper from the Monash Centre for Financial Studies has modelled retirement outcomes for a range of starting account-based pension (ABP) balances to determine the likelihood of portfolio exhaustion across various timeframes and growth/defensive asset allocations.

Authors Le and Ruthbah analysed eleven equity/bond portfolios (from 100% equities to 100% bonds) and starting balances from $100,000 to $1 million, benchmarked against ASFA's comfortable ($51,805) and modest ($32,897) standards for singles at the time of study.

The findings are stark: a retiree with just $100,000 has almost no chance of funding a comfortable lifestyle for a decade, and balances below $250,000 face a high probability of exhaustion. At the projected female median balance (~$212,000), success rates span between 74% and 89% depending on allocation, against 97% to 100% at the male median (~$283,000). Once balances approach $400,000, survivability is near-universal regardless of asset mix, lending weight to the ‘portfolio size effect’.

On portfolio design, diversified asset allocations deliver the most consistent outcomes for modest balances. All-equity strategies produce the highest expected returns but carry sharper drawdown risk, while bond-heavy allocations (despite feeling "safer") virtually guarantee accelerated capital exhaustion once mandated minimum drawdowns are applied, confirming the importance of growth assets in preserving purchasing power over the longer-term.

Sequencing risk emerges as a decisive variable: an adverse start to retirement can cut ending balances by 20% to 25%, with the effect most severe for smaller starting balances and least cushioned by asset allocation extremes. The study also quantifies a persistent gender gap; women retire with balances 20% to 30% lower than men on average, translating into materially higher depletion risk even under identical portfolio settings.

The authors conclude that “strong fund-level investment performance must be matched by reforms that address balance disparities, manage sequence risk, and ensure sustainable outcomes for vulnerable groups.”

This paper does stochastically what our own recent analysis demonstrated deterministically: sequencing risk is not a theoretical curiosity but a reaal, measurable drag on retirement outcomes.

While our April piece traced the sequencing risk effect through actual daily unit prices for two hypothetical recent retirees drawing an account-based pension through the March market volatility, Le and Ruthbah arrive at a strikingly similar finding using forward-looking capital market assumptions rather than ex-post data.

Two independent methodologies converging on the same findings should carry weight with fund CIOs and investment committees currently treating sequencing risk as a footnote to retirement portfolio design, rather than as a key drawdown-phase risk.

This is also where the paper connects directly to Treasury's Best Practice Principles, which we examined in our February special. Section 2 of the Principles directs trustees to design retirement income solutions that manage investment and drawdown risk as a core element of solution design, not an afterthought post-construction. The Monash findings give that requirement empirical teeth: a retirement income strategy that looks adequate on average-return assumptions can still fail a meaningful share of members, purely on the basis of when they happen to retire.

The gender finding deserves equal attention. A 20% to 30% balance gap compounding against sequencing exposure means women are structurally more likely to experience both triggers of depletion simultaneously; lower starting capital and higher relative sensitivity to early losses. This is not solved by retirement portfolio design alone; it points back to accumulation-phase adequacy and Age Pension interaction as the more powerful levers, together with housing equity as a potential source of uplift for homeowners.

For trustees, the RIC-lens relevance echoes our own design preferences for ABPs: time segmentation (bucketing) with intelligent rebalancing, money-weighted member returns reported alongside time-weighted figures plus drawdown rate guidance, with all three elements working in concert.

Being at the forefront of a new technology isn’t for the faint of heart, especially when that technology involves delivering complaint personal financial product advice digitally within the parameters of Australia’s financial services laws. That was certainly our experience with ‘robo advice’ over a decade ago, now rebranded as digital advice.

Which is why a new FSC report caught our attention. The paper, drawing on a nationally representative survey of 1,209 consumers (and 15 stakeholder interviews across super funds, digital providers and advice businesses), argues that Australia's advice gap is a capacity and confidence problem rather than a demand problem.

Digital tool users are consistently more likely to seek professional advice, and sooner; 28% intend to seek advice on retirement adequacy within 12 months versus 11% of non-users, rising to 44% versus 13% among digitally engaged pre-retirees. The relationship is reinforcing, not substitutive: digital engagement activates advice-seeking, while prior advice experience builds digital confidence.

The report's central framework, the Engagement Confidence Matrix, maps consumer confidence against decision consequence, arguing digital tools excel at low-to-moderate consequence decisions while human judgement remains essential as consequence rises. Only one in four digital tool users feel confident acting on a digital recommendation alone. Hybrid engagement (digital plus visible human assistance/input) is consistently preferred, especially among pre-retirees (53% of the 55–59 cohort).

On regulation, the report finds no legislative prohibition on scaled digital advice; the binding constraint is internal governance confidence and inherited conservative interpretation, not the law itself. It positions digital advice as core system infrastructure, urging super funds, providers, advice businesses, regulators and policymakers toward coordinated, proportionate execution.

This report lands squarely on the theme this newsletter has tracked across several editions: funds are moving on digital advice ahead of, not because of, regulatory certainty. Aware Super's Retirement Manager, Rest's Retire Ready and NGS Super's Virtual Adviser, amongst others, are precisely the "scoped advice deployer" and "ecosystem integrator" patterns the FSC report describes. Further, the report's finding that governance confidence, not statute, is the real constraint matches what we have observed, these rollouts proceeding without waiting on DBFO Tranche 2.

The report’s ‘Engagement Confidence Matrix’ is a genuinely useful addition to the sector's vocabulary, and it sharpens a point Brighter Super's recent findings made in our last edition: confidence and capability are not the same thing, and funds that mistake member self-reported confidence for actual retirement readiness risk under-provisioning support services exactly where and when it's needed most.

The FSC's finding that only one in four digital users feel confident acting alone on a digital recommendation is a useful corrective for any fund tempted to treat a well-built calculator as a complete retirement solution. We have long held the view that an omnichannel approach to engagement, from pure digital tools to hybrid advice to highly human-centric bespoke advice, is necessary to balance the competing tensions of cost/access versus trust. One size most definitely does not fit all, from RG 276-compliant tools and calculators to the latest AI-enabled guidance platforms.

The report's most consequential claim is structural: trustee compliance frameworks built to supervise individual adviser discretion are being applied, largely unmodified, to deterministic, auditable digital systems and that mismatch, not the Corporations Act, is what slows deployment. For trustees still treating digital advice as a bolt-on pilot rather than a governed core capability, this is the uncomfortable finding worth dwelling upon.

Margaret Cole delivered her farewell remarks as APRA Deputy Chair at a Conexus boardroom lunch on 25 June 2026, closing a five-year term that began on the same day Australia's Your Future, Your Super legislation took effect. Reviewing the arc of her speeches since 2021 traces the industry's gradual shift from an accumulation-centric focus toward genuine retirement awareness and outcome delivery.

In her earliest public remarks on the covenant, Cole framed the opportunity ahead of the obligation, telling trustees the challenge was about helping members transition with confidence beyond the accumulation phase. That optimism didn't survive first contact with the data. By August 2023, presenting APRA and ASIC's joint thematic review alongside ASIC's Jane Eccleston, she was more direct, pointing to the lack of progress and urgency demonstrated by some trustees in implementing their strategies. Weeks later, in a speech she titled ‘Fortune favours the brave’, she repeated the diagnosis in sharper terms: trustees were improving, but not with the urgency required.

At the 2024 Conexus Financial Superannuation Chair Forum, Cole flagged that “high on our agenda is our continued push with ASIC into improving retirement income”. At an August Conexus Retirement Conference address she described the covenant's 2022 introduction as a ‘clarion call’ for trustees to shift their attention to improving retirement outcomes, one that was yet still not fully answered. Progress remained mostly incremental: a mid-2024 pulse check found just one in five planned trustee improvements had been completed in good time.

By 2025, addressing the FSC's Innovation in Retirement conference, the tone had shifted from admonishment toward infrastructure-building; APRA's push to integrate retirement product data into its Comprehensive Product Performance Package and work with Treasury on the forthcoming Retirement Reporting Framework, all in service of what Cole saw as the goal of improved outcomes for superannuation members.

Her farewell remarks close the loop. Reflecting on transparency as her era's defining reform, Cole described her mission as leaving the super system fitter for the future, "more closely attuned to the needs of the members, ready to fulfil the commitment of delivering on the promise of dignified retirement." She left trustees, regulators and industry with a final challenge: to be "drivers of a stronger system and of better outcomes for real people", a throughline connecting her first speech on the covenant to her last.

Read end to end, the immediate past Deputy Chair’s APRA speeches form less a farewell tribute than a running scorecard; one that isn’t flattering. Four years since the covenant took effect, the key take-outs from APRA’s joint 2023 review with ASIC through to her 2026 final remarks are remarkably consistent: trustees improving, but not fast enough, not thoroughly enough, and with insufficient grounding in genuine member data.

The mid-2024 finding that only one in five planned improvements had been completed by the pulse-check deadline is the number that should be most concerning. That wasn't a baseline assessment catching an industry at the starting line, it was a follow-up review, measuring trustees against commitments they themselves had made in response to being told, the year prior, that their strategies lacked urgency. A 20% completion rate against self-set deadlines is not a data or resourcing problem; it is a prioritisation problem.

What didn't change across five years of regulatory pressure? Gaps in understanding member needs, variable data quality, and thin cohort-level design. And per our coverage of the Corrs Chambers Westgarth review of public-facing retirement income strategies in the last edition, they remain relevant in 2026; just now against Treasury's Best Practice Principles rather than the covenant's original intent.

Margaret Cole's departure raises a genuine continuity question. Five years of consistent, patient pressure from one dogged regulator built a track record trustees could set their expectations against. Whether her successor maintains that cadence, or whether transition creates a pause the industry reads as reduced urgency, will be the first real test of whether the gap she leaves behind starts closing or simply persists under a new watch.