Scott Heathwood APIG 2026 Address
WHO PAYS WHEN THE CHAIN FAILS Address to the APIG National Conference Scott Heathwood, President and Chairman, Institute of Financial Professionals Australia Sofitel Sydney Wentworth, Thursday 10 September 2026 PART 1 · THE ADDRESS Shield and First Guardian was not a marketing…
WHO PAYS WHEN THE CHAIN FAILS
Address to the APIG National Conference
Scott Heathwood, President and Chairman, Institute of Financial Professionals Australia
Sofitel Sydney Wentworth, Thursday 10 September 2026
PART 1 · THE ADDRESS
Shield and First Guardian was not a marketing failure. It was not a breakdown in lead generation. It was an integrated scam, and it worked because the people who built it did not start by tricking clients.
They started by tricking the gatekeepers.
The people whose job it is to say yes or no to whether a product even makes the shelf. The trustee who decides what goes on the platform. The research house that rates it. The responsible entity that is supposed to be running the scheme for its members rather than for itself.
By the time anybody answered an advertisement about their superannuation, every one of those doors had already been opened.
Woolworths does a better job of deciding what goes on its shelves than some of these trustees did.
And not one of those omissions was priced. Not by the regulator, not by the market, and not, I would gently say to this room, by you. Which is why the whole thing now sits like a black cloud over the advice sector that for the most part were never involved.
So I want to put 2 batches of questions to you, because everything I say after this turns on the difference between them.
The first series of questions to be asked is:
Who are the sponsors and will this product do what the disclosure document says it will do?
Ie; Will the money go where the document says the money goes?
And who, exactly, is enforcing that?
The second.
Is that product right for this client, sitting in front of me, on their circumstances, today.
Those are different questions. Different people answer them, at different points in the chain, for very different money. One is answered once and reaches everybody. The other is answered thousands of times and reaches 1 family at a time.
And we have built an entire liability system around the second one, and almost nothing around the first.
I have spent the past year looking at Shield and First Guardian. What strikes me is not that people broke the rules. It is that most of them did not.
There is a reason for that, and it is not a comfortable one. In this country conflicts of interest are not prohibited. They are permitted, so long as the members' interests are prioritised and the members give their informed consent. Trust law prohibited them outright. Corporations law does not.
Dr David Millhouse, a legal scholar at Bond University, put it plainly in the Financial Review in February. It is open to the directors, he says, to determine what the statutory best interests duty under section 601FC(1)(c) actually means, and whether they have exercised it rather than their own self-interest.
Read that twice. The people holding the conflict decide what the duty requires of them.
So here is what happened. Around 11,000 Australians put approximately $1 billion into those 2 funds. Most began by answering an advertisement about their superannuation. They gave a telephone number to a call centre, and some were passed to an adviser authorised by an AFSL. The funds sat on a platform run by a licensed trustee, who took its fees, and a responsible entity ran the scheme underneath. And every one of those agents relied on research from a known and reputable research house.
There were 12 significant relationships in that chain. On my own count 11 of them carried a conflict of interest, and you need not take my count on trust, because the detail is on the public record.
The 2 men who founded the scheme's responsible entity were also directors of its investment manager. As trustees they were guarding investors against the actions of themselves as fund managers. The property arm invested into a wholesale fund managed by a company one of those men controlled, and 9 of the 11 projects it financed, book value $214 million, were controlled by him or by companies he directed. The only independent member of the investment committee lasted about a year. His successor says he was never a member at all, despite his name appearing in a product disclosure statement and in a research report. And the research house that rated the fund says it did not know who controlled the properties underneath it.
Related party disclosures were never made to retail investors. And almost none of it required anybody to break a law.
Now here is the part I think has been missed, because it is not about the scheme at all. It is about why anybody answered that advertisement.
Nobody lifted the sheets. Nobody noticed thousands of investors flocking to a relatively new product. Nobody asked why their interest had been piqued at a time of record complaints against the superannuation sector, with AFCA having just had its busiest year on record and ASIC delivering a scathing review of how funds handle death benefit claims. Nobody connected it to the hundreds of thousands of members compulsorily moved out of options they had chosen into defaults they had not. Nobody noticed it began after the family accountant had been removed from giving simple advice well within their wheelhouse. And nobody remembered that Hayne had just shown the country that the banks and AMP were bad actors in this sector.
And the trustees of the major institutions were only too pleased to facilitate a migration of funds out of the industry sector and onto their own platforms.
And the advisers, many of whom were salaried employees of the people running the scheme, were told to get on with selling it, because it was supported by the research and by big names lending their reputation to the whole process.
That is the failure. An omission to question the moving parts. If anybody was required to ask, they were told a lie. If they were not required to ask, nobody asked.
Before I come to who pays for that, I want to show you who gets paid.
In July I sat in a room with 5 other advice licensees. Between us, 936 advisers and $48.2 billion under advice. We put our actual accounts for last financial year on the wall and added them up. Not a survey, not a model. 6 sets of real numbers.
The advice relationships in those 6 businesses generated $741,531,000 of revenue across the financial services chain.
$416,631,000 of it went to platforms, investment managers and insurers. That money never appears in an advice business's accounts. We do not invoice it. We never see it.
$324,900,000 arrived. Of that, $283,355,000 went to the advisers themselves and $30,503,000 went to operating costs, which is licensing, compliance, technology and professional indemnity.
$11,040,000 was left. 6 licensees. 936 advisers. A year's work. $11,040,000.
Hold the shape of that in your mind. It narrows at exactly the point where the liability lands.
Vertical axis, what the market pays for each segment as a multiple of earnings. Horizontal axis, measured risk, built from AFCA complaint data and from Lloyd's underwriting syndicates. That is your own industry's view of where the risk sits, not mine.
Platforms carry the lowest measured risk in the chain, 0.3, and they trade at about 95 times earnings. Listed advice businesses, on the same exchange on the same day and on the same measure, trade at about 11 times. And the advisory networks carry the highest measured risk of anybody, 2.3.
95 times earnings for the platform. 11 times for the listed advice business. Same market, same measure, and the risk runs the opposite way.
I used to call that the market pricing revenue rather than risk. I no longer think that is right. If it were pricing revenue, the platforms would carry the heaviest rating in the chain. They take the largest share of it. They carry the lightest.
Here is what I think is going on. A platform trustee does 2 entirely different jobs. It runs an administrative business, onboarding product and collecting basis points, which is stable, scalable and detached from the member. It also holds a fiduciary office. It decides what goes on the menu, and it owes the members their interests ahead of anybody else's.
2 duties, different in kind and different in consequence, and the market has collapsed them into a single number that prices the first and ignores the second. That is a bifurcation the marketplace appears to have forgotten about.
And then Shield happened, and for once the money came out of the fiduciary limb rather than out of adviser professional indemnity. Macquarie paid about $321 million. Netwealth paid more than $100 million. Out of a segment your own market rates at 0.3.
The 2 questions, and who answers them
Let me put the trustee's job next to the adviser's, because I do not think the difference is properly understood, even inside our own industry.
Start with access. If a product is not on a platform it is effectively unavailable. A licensee will not put it on an approved product list, because an adviser cannot reach it at a retail level. So the trustee is not 1 option among many. The trustee is the gate. Almost nothing reaches a retail client in this country without one of them saying yes.
And when they say yes, they are not merely admitting a product to a menu. They are lending it their brand, and that is a tacit endorsement whether they intend it as one or not, because it is precisely what the adviser and the client take it to be. They cannot pretend otherwise. It is a simple fact of the function, and they are handsomely rewarded for performing it, as the last slide showed you.
That judgment is made once. Not client by client. It is made in the abstract, for everybody, and it either holds or it does not.
Now the adviser. The adviser's duty does not sit at the client base level and it does not sit at the funds under management level. It sits at the individual file. Every single file. Is this product right for this person, on what they have told me about their circumstances, their objectives, and what they can afford to lose.
Those are not the same test. They are not even the same kind of test.
The trustee has to decide whether a product is sound at all. The adviser has to decide whether a sound product suits 1 particular person. And the adviser's question presupposes the answer to the trustee's. The adviser is not assessing whether the product is generally acceptable. That has supposedly already been done, by the trustee, with a research house rating on top of it, on a platform whose operator has taken a very large share of the value in this chain and might reasonably be assumed to have looked at the promoters and sponsors it is letting through its own technology.
Now take the adviser who recommended a Shield product to a client who said they wanted some property exposure and did not mind a bit of development risk. On the evidence in front of that adviser, a named platform, a licensed trustee, a rating from a leading property research house, that recommendation was defensible. At face value it was a suitable product for that client.
I do not necessarily accept that it was a failure at the file level but a lot was,
It was a recommendation made on a foundation that had already collapsed before the adviser ever saw it. And that is an entirely different thing from failing to notice that the people running the scheme sat on both sides of the fence, related to one another by ownership, by directorship and by commerce. The lender. The developer. The development manager. Everybody in the food chain.
That is where the adviser is blind. Not negligent. Blind. And it is the 1 thing nobody has priced.
And here is why that matters more than anything else I will say today. Get the file level question wrong and you hurt your own client. Get the product level question wrong and you hurt everybody who was ever going to be offered it.
11,000 people. 1 decision, made once, at the gate. That is where the systemic risk in this industry actually sits, and it is the only part of the chain that is not rated, not loaded, not excluded and not levied.
So where is the layer that is supposed to answer for this?
$1.1 billion went into those funds. Behind a registered scheme of that size a responsible entity must hold $5 million of capital, and above that cap the requirement stops rising altogether. An entity running $10 billion holds the same $5 million.
You do not have to take my word for what that capital is for. ASIC said it plainly in March.
The financial requirements for responsible entities are not intended to address market or credit risks, or to prevent entities from becoming insolvent or failing. They are also not designed to compensate clients for unexpected losses.
Not designed to compensate clients for unexpected losses. That is the regulator, about the layer directly above the adviser.
So this is where the bill lands. On the narrow end.
There is a cap in the legislation. $20 million is the most the advice sector can be levied for the Compensation Scheme of Last Resort in a year. The estimate for the coming year is $190.3 million. And the entire professional indemnity premium pool for financial advice in this country, everything underwritten, rated, reserved and reinsured, is in the order of $100 million.
So the layer nobody prices is roughly twice the size of the layer you do.
$416 million left that chain at the top and never troubled an advice business's accounts. The bill arrived at the end that had $11 million.
Value in this chain accrues to whatever can be scaled. Liability accrues to whatever cannot.
Now the part I actually came here to say
None of this is new, and that is what I want to leave you with.
Storm Financial, January 2009. 13,000 clients. $4.5 billion under advice at its peak. A model built on borrowing against the family home and borrowing again against the portfolio. The Commonwealth Bank held about 30% of the margin loan book.
Asked about it by a parliamentary committee, the chief executive of that bank said the local relationship with Storm was sometimes too close and that on occasion they lost objectivity. He also said it was not a relationship that ran to the highest levels of the bank.
There it is again. Nobody at the top looked.
Then look at what we did about it. After Storm came the Ripoll inquiry, then FOFA. Then MySuper, legislated by one government and rolled out rather than repealed by the next. In 2016 we removed the accountants' exemption. Then the Royal Commission. Then we cancelled insurance inside superannuation for anybody with a balance under $6,000 unless they wrote in and asked to keep it, a catastrophe for those who had become uninsurable and could not be reached. Then Dixon Advisory, and the compensation scheme we are now arguing about. And now Shield and First Guardian.
I will put it more strongly than most people in my position would. I think every significant regulatory response of the last 25 years has made matters worse. Not one of them made a single person responsible for an outcome.
And the pattern runs further back than my memory of it. Millhouse examined 199 senior court judgments between 1984 and 2018 and found conflicts of interest a common feature of every major collapse of a non-bank financial entity in this country across that whole period. He estimates the investor funds lost or impaired at around $52 billion.
$52 billion. 34 years. The same cause every time.
My concern is that we are about to do it again. The instinct after an event like Shield is well established. Re-rate the adviser. Load the excess. Write an exclusion. Add a levy. We have run that experiment for 17 years, it has not worked once, and I would put it to you that it made things worse.
After the Royal Commission cover became scarce, dear and narrow, and adviser numbers fell by roughly half. Advice quietly became a product for people who already had money. Which is where the 11,000 came from. Not from established advice firms. From cold calls aimed at people with modest balances and nobody left to ring.
Withdrawing capacity did not remove the risk. It moved the risk to where nobody was watching.
And here is the clearest example of why doing it again will not work. Most policies in this market exclude, or heavily qualify, claims arising from advice on product that is not on the licensee's approved product list. Shield and First Guardian were on approved product lists. They carried research house ratings. They sat on regulated platforms, accepted by licensed trustees. Every box was ticked.
So that exclusion would not have excluded 1 dollar of these claims. Meanwhile it bites on the adviser who went off the list for a considered, documented, client specific reason. It excludes the thoughtful deviation and admits the catastrophic consensus.
What has to change
Somebody has to own the outcome. That is what all of this comes down to.
There is something else in it too. We tell the public we are going to fix it. We tell them there is a scheme, a regulator, a soft landing. What we have stopped telling them is to take an interest in their own affairs. People leave their rubbish in the park because they know the council will pick it up. If somebody else is going to clean up after you, you stop being careful.
We are not fixing this. We are administering it.
So, 4 changes. 2 of them need Canberra. 2 of them this room could start on Monday.
One. Underwrite the chain, not just the firm. Cover that attaches to the scheme, is sized to the scheme property, and survives the liquidation of the entity that caused the loss. Every compensation mechanism in this chain dies with the entity that caused the damage, and that single design fault is why the Compensation Scheme of Last Resort exists.
Two. Price the gatekeepers in the middle, because that is where a bad product dies cheapest.
This is the one I would press hardest. Think about where Shield could have been stopped. 1 research house declining to rate it stops it reaching thousands of advisers. 1 platform trustee refusing to onboard it stops it reaching every adviser on that platform. 1 responsible entity asking where the money actually went stops it in its first year.
So rate them on what they let through. Make independent confirmation that the money went where the product disclosure statement said it would an insured and audited condition of cover. A research house or a trustee whose own premium turns on the quality of what it waves past will look properly, and it will look before the damage is done.
Three. Let AFCA join the whole chain to a complaint, and apportion between it.
The consumer complains about the adviser, because the adviser is the only person they ever met. AFCA can only deal with its own members. And the courts will not apportion either. In 2015 the High Court held, in a case called Selig against Wealthsure, that proportionate liability applies only where a claim is confined to misleading conduct. Plead anything else for the same loss, negligence, breach of the best interests duty, breach of contract, and there is no apportionment at all. So a plaintiff simply adds a second cause of action and the whole regime disappears, which means the last solvent party carries 100% of it. That party is the licensee.
Give AFCA the power to join every party in the chain and apportion between them on the facts, and make membership follow the chain so the parties who caused the loss are within its reach. That alone would cut the cost falling on advisers, and put the pressure where the conduct was.
Four. Close the fitness for purpose gap.
Buy a toaster in this country and it must be fit for purpose. That is a statutory guarantee and the manufacturer wears it. Buy a managed investment scheme and there is no such guarantee, because financial products are carved out of the Australian Consumer Law by section 131A of the Competition and Consumer Act. The mirror provision in the ASIC Act covers services rather than products, and excuses the supplier where it would be unreasonable for the consumer to rely on their judgment.
If a fee is taken for a product that was never fit for purpose, that ought to be a consumer law problem for the people who built it and waved it through, not a best interests problem for the last person to touch it.
I will finish where I started, with the 2 questions.
One of them is systemic. It is answered once, at the gate, by the best paid party in the chain, and it reaches everybody. And it is not priced, not insured against, not apportioned, and on the evidence of the last 17 years not much looked at either.
The other is answered 1 file at a time, by the least paid party in the chain, and it reaches 1 family. And that is the one we price, exclude, levy and litigate.
Until those 2 are the other way round, we will be back in this room in 5 years, talking about a different product with a different name.
