The $327,800 gap: how Australia's super industry learned to lie with data

On a Wednesday morning in late February 2026, the Super Members Council dropped a dataset that was designed to frighten people. Australians were abandoning their default superannuation funds at an accelerating rate, up 17 per cent year-on-year, and the people doing the switching were, according to the SMC’s analysis, overwhelmingly young, financially vulnerable, and operating without professional guidance.

The average balance of someone switching to a superannuation platform, the SMC announced, was just $21,700. Seven in ten of those switchers had no prior financial advice relationship. The implied conclusion was barely disguised: something predatory was happening inside the ‘choice’ superannuation sector, and it needed to be stopped.

The data landed on the desks of journalists, politicians and policy advisers simultaneously. The ABC ran the story that afternoon. By the following morning, the narrative had taken shape: unsophisticated consumers, stripped of the protection of large default funds, were being funnelled into expensive and risky structures by operators who did not have their interests at heart. The failures of Shield Master Fund and First Guardian, which had recently crystallised into genuine consumer losses for approximately 11,000 Australians who had invested around $1.1 billion combined, provided the atmospheric backdrop that made the story almost write itself.

Less than four weeks later, the Financial Services Council published its rebuttal. The average person opening a new superannuation platform account, the FSC announced, was 59 years old. Their average balance was $349,500. Almost all of them were working with a professional financial adviser. Even the youngest cohort, under-30s using platforms, averaged $54,800, more than three times the $16,200 average balance of their peers in APRA-regulated funds. Those aged 30 to 45 averaged $162,300, against $80,800 for their peer group in the broader system. The choice sector, the FSC declared, was not preying on the vulnerable. It was serving the sophisticated. The SMC’s claims were not just wrong, but ‘highly misleading.’

Both statements were backed by data. Both were delivered by credible peak body organisations. Both were presented to the media as evidence.

And both, in ways that matter enormously to sound policy and to the long-term credibility of the industry, were conclusions that had been written before a single spreadsheet was opened.

The arithmetic of convenient truth

Before examining how each organisation reached its preferred destination, it is worth pausing on the number that sits at the centre of this debate and refuses to be explained away.

This is not a rounding discrepancy. It is not a definitional nuance. It is a chasm of such magnitude that it can only exist if the two organisations were studying fundamentally different populations, and presenting both, to a media and policy audience without the time or inclination to read methodology notes, as representative of the same phenomenon.

They were. And both knew it.

The SMC drew its data from five large industry funds, funds that were not named in the press release, supplemented by ATO and APRA data. The primary sample measured only the members who left those specific funds and moved to platforms. The FSC drew its data from the seven largest platform providers by funds under administration, the biggest players in the wrap and managed account market by the most generous possible weighting metric. The study measured only the clients who arrived at those specific platforms.

These are not two methodologies measuring the same thing from different angles. They are two research instruments designed to observe different populations, both claiming the authority to describe a single phenomenon. The SMC’s telescope is pointed at people leaving large industry funds. The FSC’s telescope is pointed at people arriving at the most prestigious and largest platforms in the country. Naturally, they see different things. The intellectual dishonesty is in presenting those different observations as contradictory evidence rather than as complementary, and equally partial, views of a complex landscape.

The unnamed five

The decision by the SMC not to name its five source funds is, in isolation, a methodological failure serious enough to invalidate the research’s use in public policy advocacy. In a peer-reviewed academic context, unnamed sources in a study of this nature would not clear an editorial review process. In a government submission, it would attract explicit scrutiny. In a peak-body press release distributed to national media and used to call for regulatory intervention, it passed largely without comment.

This matters because the choice of which funds to include in the sample is the single most consequential methodological decision in the entire study. The departing member profile of a fund like AustralianSuper, with its enormous, diversified membership base spanning every income bracket and life stage, looks structurally different from a sector-specific fund with a concentrated membership profile. A fund that has recently experienced publicised governance concerns will lose a different profile of departing member than a fund with an unblemished record. The identity of the five source funds is not a footnote: it is the study.

The decision to withhold that identity is, at minimum, a failure of research transparency. At maximum, it suggests the organisations involved understood that naming the funds would enable independent analysts to interrogate whether the sample was genuinely representative, or whether it was selected precisely because it produced the desired distributional outcome.

The SMC is not alone in this practice. The financial services industry has a long and undistinguished history of commissioning research that is technically defensible and practically engineered. Name the sample, specify the measurement period, choose the benchmark, weight the methodology and you can produce almost any finding you need. The unnamed five is simply the most recent and most visible example of a technique that has become standard operating procedure in peak-body advocacy.

The FUA weighting problem

The FSC’s methodology deserves equal scrutiny, and receives rather less of it in the coverage of this dispute.

Weighting platform data by funds under administration is, on its face, a reasonable approach to ensuring that the largest and most significant operators contribute proportionately to a dataset. In the context of this specific research question, ‘what does the average platform switcher look like,’ it is a choice that systematically biases the sample toward exactly the population the FSC needs to find.

The platforms that dominate the Australian market by FUA, HUB24, Netwealth, Praemium and their institutional-grade peers, built their businesses over twenty years of deliberate distribution strategy. They cultivated relationships with the most productive, most credentialed, most high-net-worth-focused advice practices in the country. Their average client is wealthy, older and professionally advised not because platforms as a category attract that profile, but because those specific platforms were architecturally designed and commercially incentivised to attract exactly that profile.

Weighting by FUA means these platforms dominate the FSC’s sample. A platform with $40 billion in FUA and 50,000 high-balance clients contributes twenty times as much to the dataset as a platform with $2 billion in FUA and a different client mix, even if the smaller platform is far more relevant to the policy question being examined, because it is closer in profile to the operators whose governance frameworks have actually been implicated in consumer harm.

The research is not false. It is simply answering a question that nobody asked, with great methodological confidence.

The advice attribution gap

Of all the competing claims in this dispute, the question of advice relationships is the one that most rewards careful analysis, and most punishes the casual reader who accepts either side’s framing.

The SMC’s finding that 70 per cent of platform switchers had “no prior advice relationship” was the data point that generated the most alarmed commentary. It suggested a large population of financially unsophisticated consumers making consequential superannuation decisions without professional guidance. In the context of recent platform failures, it was genuinely concerning.

The methodology behind the number, however, is deeply problematic. An “advice relationship,” as measured from a large industry fund’s administrative data, typically means one specific thing: a financial adviser who has lodged paperwork through that fund, an adviser service fee arrangement, a rollover authority, a binding nomination prepared through the fund’s own advice infrastructure. It is a measurement of advice relationships that are visible to the departing fund’s records. It is not a measurement of advice relationships that exist.

An individual who sought advice from an independent financial planner, paid a fee-for-service, received a detailed Statement of Advice recommending a platform rollover for reasons of investment flexibility, estate planning efficiency or retirement income structuring, and then executed that rollover, would appear in the SMC’s dataset as “unadvised.” Not because they received no advice. Because the advice was delivered and recorded outside the departing fund’s administrative architecture.

The FSC made this point directly, with CEO Blake Briggs noting that “default funds cannot see a member’s full financial position, including whether multiple accounts are being consolidated over time or whether an adviser is using a platform to manage the household wealth of several family members under a holistic strategy.” The FSC is correct on the analytical point. The gap between “no advice relationship recorded in this fund’s systems” and “no advice relationship” is the difference between “these people were exploited by unscrupulous operators” and “these people received advice that is invisible to the measurement instrument being used.”

The FSC’s counter-claim, that “almost all” platform users are advised, has its own circularity. Almost all clients of the seven largest platforms are advised because those platforms are predominantly distributed through professional adviser channels. They were built that way. It is worth noting that independent research from CoreData found approximately 70 per cent of super switches are adviser-driven, which provides partial directional corroboration of the FSC’s claim, though it does not validate the specific sample methodology. Measuring the FSC’s seven and concluding that platform users are almost universally advised is not a finding. It is a description of the commercial architecture of the sample.

The genuine question, what proportion of all platform-switchers across all platform types, including smaller and less established operators, received genuine, quality-assured financial advice before switching, remains entirely unanswered by either dataset. This is the question that policy needs to answer. Both sides have chosen, for different reasons, not to answer it.

The ghost in the room

Hovering over every paragraph of this debate, though rarely named directly by either party, is the actual evidence of concentrated consumer harm: Shield Master Fund and First Guardian.

These failures, real collapses with real losses borne by approximately 11,000 real people who had invested around $1.1 billion, are the legitimate policy concern that should be anchoring this conversation. They are invoked atmospherically by both sides: the SMC uses them as backdrop evidence that the choice sector is dangerous; the FSC dismisses them as ‘edge’ cases that should not contaminate the policy environment for the well-governed majority. Both responses are inadequate, and both obscure a factual complexity that is material to the policy debate.

The FSC’s framing, that the major institutional platforms were categorically separate from the Shield and First Guardian failures, does not survive contact with the enforcement record. Netwealth, one of the largest platforms in the country by FUA and almost certainly present in the FSC’s seven-platform sample, admitted to failures relating to First Guardian and agreed to pay more than $100 million in compensation to more than 1,000 affected members. Macquarie was ordered to pay $321 million to Shield investors. Equity Trustees, as trustee for platforms including Super Simplifier, faces civil proceedings from ASIC over its role in making Shield available as an investment option. These are not peripheral actors: they are mainstream institutional participants whose governance failures contributed directly to the harm the FSC characterises as a problem confined to the margins.

This does not validate the SMC’s conflation of all platforms as equivalently dangerous. The governance failures that drove Shield and First Guardian were concentrated in specific distribution practices, involving lead-generators, unscrupulous advice firms, and inadequate trustee oversight, rather than in institutional-grade wrap infrastructure per se. The FSC is correct that this distinction matters. It is less forthcoming about the fact that some of its own members were found wanting precisely when that trustee oversight was tested.

If this debate were genuinely about consumer protection, both sides would be engaged in the analytically difficult work of characterising what exactly went wrong with those specific operators, and which elements of that failure were structural rather than individual. Neither is. The SMC is using the failures rhetorically. The FSC is minimising them defensively, even as its own members write nine-figure cheques. The consumers who lost money are, in the policy conversation at least, mostly absent.

A thirty-year pattern

It would be convenient to treat this episode as an isolated outbreak, two organisations behaving opportunistically in the context of an unusually heated policy dispute. The history of Australian financial services does not support that comfort.

The industry has been here before, repeatedly and with depressing consistency. In the lead-up to the Future of Financial Advice reforms, both sides produced research that confirmed what they already believed about the value and cost of professional advice. In the post-Hayne Royal Commission debate over vertical integration, both institutional defenders and reform advocates generated data that mapped precisely onto their predetermined conclusions. In the ongoing superannuation performance benchmarking debate, the choice of benchmark, measurement period and fee treatment each carry enormous analytical weight, and each is chosen by people who are not, and cannot reasonably be expected to be, indifferent to the outcome.

The pattern is structural, not incidental: identify the policy conclusion you need, design a sample that produces it, present the result with the authority of research rather than the transparency of advocacy, and distribute it through media channels before the methodology receives independent scrutiny. The approach works in the short term. It shapes political conversations, influences regulatory outcomes and ‘wins’ news cycles. In the long term, it is dismantling something the industry needs desperately and is losing steadily: the presumption of good faith.

The Hayne Royal Commission did not happen because individual operators made individual mistakes. It happened because a cumulative pattern of behaviour, of institutions using their information advantages, their research capabilities and their policy access to serve themselves while performing consumer focus, eventually became too visible to ignore. The lesson was supposed to be that transparency is not a strategic vulnerability. It is a survival requirement in an environment where trust, once lost, is extraordinarily difficult to rebuild.

What evidence would actually look like

The requirements for research that could genuinely inform policy on this question are not technically sophisticated. They are simply inconvenient for organisations whose research budgets are allocated to producing conclusions, not discovering them.

1: a common data standard

Agreed definitions of “switcher,” “advised,” “platform,” and “balance at point of switching” were established before data collection began, not reverse-engineered to produce the desired comparison. APRA’s recently finalised Retirement Reporting Framework, now in implementation consultation with final reporting standards targeted for Q3 2026 and first data collection not before Q4 2027, is a step in this direction. It will not arrive quickly enough for the current debate.

2: named and auditable sources

Any dataset used to advocate for regulatory change affecting the retirement savings of millions of Australians should fully disclose its sources, sampling frame and material selection biases. An unnamed member fund in a peak-body press release is not research. It is a claim with a confidence interval attached.

3: distribution analysis over means

The mean balance of platform-switchers tells you almost nothing policy-relevant, because the harm, if it exists, is concentrated in the lower quartiles of the distribution, not averaged across the whole population. Show the full distribution. Show the median. Show the bottom and top deciles. Then have the conversation.

4: categorical separation of platform types

The failure to distinguish between major APRA-regulated wrap platforms with mature governance frameworks and smaller, less-established operators is analytically indefensible given that the entire policy debate is ostensibly about harm concentrated in the latter. Any research that treats all platforms as a homogeneous category is not equipped to answer the question it claims to address.

5: independent verification

Research used to shape national retirement savings policy should be independently reviewed before it becomes the basis for public claims. Both the SMC and FSC have the resources to commission this. The choice not to do is a choice.

The real cost

The people who lose most from this pattern are not the FSC or the SMC. Both organisations will survive this dispute, publish more press releases, and continue representing their members with vigour. The people who lose are the financial advisers trying to have an honest conversation with a client about whether a platform rollover is appropriate for their circumstances, a conversation that now takes place in an environment saturated with competing narratives, none of them disinterested. The people who lose are the consumers who cannot distinguish between a governance-grade wrap platform and the kind of structure that costs people their retirement savings, because the people who should be helping them make that distinction are busy fighting a data war. The people who lose are the policymakers who need accurate, disinterested evidence to design regulatory frameworks that are actually targeted at the actual harm.

And the people who lose, ultimately, are the member funds and the platform operators themselves, because an industry that cannot produce credible, disinterested research is an industry that is writing the case for external intervention on its own behalf.

The answer to the genuine policy question, ‘how do we protect consumers from the specific harm vectors demonstrated by Shield and First Guardian without restricting the legitimate exercise of informed consumer choice,’ is not complicated in principle. It requires distinguishing between operators who caused harm and those who did not. It requires targeting regulatory interventions at specific governance failures, not broad structural categories. And it requires research that was designed to find the truth rather than confirm it.

The $327,800 gap between the FSC’s average and the SMC’s average is not a statistical curiosity. It is the precise measurement of how far Australian financial services is willing to stretch the meaning of the word “evidence” when the stakes are high enough.

That gap is the story. Everything else is methodology.

Benjamin Walsh is the founder and Principal Consultant at WealthVantage Partners (ABN 98 282 899 515), a boutique research and strategy consultancy specialising in wealth management, financial advice, superannuation, and retirement income. He has thirty years of experience across investment management, wealth management, and financial advice, and holds postgraduate qualifications in finance and wealth management.

This article represents the analytical views of the author in their capacity as an independent commentator. It does not constitute financial, legal, or regulatory advice. All factual claims are sourced from publicly available primary sources current as at March 2026.