When Super Outgrows Its Market
Australia’s superannuation system is on a path beyond A$6 trillion. For its largest funds, the harder question is how to turn an active idea into a position large enough to matter without losing its edge.
By the team at Banyantree Investment Group
This essay is adapted from analysis first shared in our Monthly Investment Letter of 21 July 2026.
One per cent of Australia’s superannuation system is about A$45 billion. On the local exchange, that is the scale of one of its largest companies.
The system held A$4.49 trillion at the end of 2025.¹ The entire Australian share market, every entity quoted on the ASX, was worth A$3.26 trillion at 30 June.² The two figures measure different things, and the comparison is usually made carelessly. Super owns bonds, property, infrastructure, private assets and a growing share of everything outside Australia, and its funds do not act as one portfolio. But set beside each other, the numbers size the task. The national savings pool is closing in on the value of the entire market it grew up investing through, and it is still compounding.
The demand story writes itself, which is why it is the one usually told. Projections published this year put the system on a path past A$6 trillion by 2035.³ Compulsory contributions arrive every payday, regardless of price or mood. Global managers can read that arithmetic as well as anyone. The first-level questions all live inside that story: which assets benefit, whether the flow puts a floor under the local market, how to win a share of the fee pool.
Those are reasonable questions. We think the more useful one runs the other way: what the market can still do for the pool. How much differentiated return can a buyer this size still pursue, and what happens to the strategies it can no longer fit through the door?
Start with stock and flow, because the flow gets the attention and the stock does the work. Contributions into the system ran to A$220.8 billion in the year to December 2025, against A$139.9 billion paid out in benefits.¹ Call the organic inflow A$80 billion a year: persistent, contracted, growing with wages. Now move markets 2% in either direction. That is almost A$90 billion, more than the year’s entire organic inflow, delivered in an afternoon. The pool’s size is made of past returns far more than future contributions, which is one reason A$4.49 trillion is no floor beneath Australian asset prices. The flow is real. It is just smaller than the thing it is supposed to be holding up.
And the system will not run out of things to own. The naive version of the capacity worry, Australia somehow running out of assets, is wrong and easy to knock down. Broad market exposure scales almost without limit; an index behaves the same way at A$4 billion as at A$400 billion. What does not scale is the other thing funds are paid to do. Capacity is the word everyone in this industry uses and almost nobody defines, so define it properly: the amount of a security that can be held at a size that matters, entered and exited on terms that leave the idea intact. On that definition the system has effectively unlimited capacity for the market’s return, and a shrinking capacity for anything different from it.
The constraint bites at two levels, and it pays to keep them separate. An active position has to be large enough to matter within the strategy taking the risk. And the strategy has to be large enough to matter to the fund allocating to it. A pool can clear the first test and fail the second, and as the pool grows, the failure point climbs.
Run it the way a portfolio construction meeting would. Take a fund the size of the country’s largest, roughly A$400 billion today and, on that fund’s own published forecast, past A$600 billion by 2030.⁴ Suppose it finds a genuinely mispriced company worth A$2 billion, comfortably inside the ASX 200, and wants the idea to show up in members’ returns. A position worth 0.10% of the fund, near the floor of what can influence a total-fund result, requires A$400 million. That is 20% of the company, awkward on its own and, as it happens, exactly where Australian takeover law draws its line: a buyer can reach 20%, but moving above it requires one of the permitted routes under the takeover provisions.⁵ The position the arithmetic asks for sits at the very edge of what the law lets a shareholder build.
So suppose it settles for half: A$200 million, or 10% of the company. The shares might trade A$5 million a day, a normal figure at that size. Taking a fifth of daily volume, which is aggressive for a buyer trying not to move the price, the fund absorbs A$1 million a day, and the position takes 200 trading days to build. Ten months of continuous buying, disclosed to the market in substantial-holder notices once the stake passes 5%,⁵ for a holding worth five basis points of the portfolio. If the thesis is right, the result is barely visible in the fund’s return. If the thesis is wrong, the exit takes longer than the entry, into whatever bid remains once the fund’s own selling is the news. Vary the inputs and the months stretch or shrink. The shape of the problem doesn’t.
So the meeting ends with four doors. Force the position anyway and wear what that takes. Hand the idea to a specialist manager. Leave the market for one with bigger companies in it. Or hold the company at its benchmark weight and let the mispricing sit there, owned and unexploited. The system’s listed-market response to its own scale runs through those four doors, so take them in turn.
The first door has a public exhibit: a fund that decided a position should matter, and showed everyone what that now involves. The episode is sometimes read as proof that scale still works in listed markets. Read the mechanics and it shows the opposite: what it now costs, and what a position must become, for size to matter at all. Through late 2023, AustralianSuper built its holding in Origin Energy to about 17% while a Brookfield-led consortium was trying to buy the company for about A$20 billion, said publicly that the offer sat below its view of long-term value, and voted the scheme down that December.⁶ Whatever one makes of the judgement, notice what the position had become. At that scale the stake had stopped being a view about price and become a veto over outcomes. The act of building the holding changed what could happen to the company, which is the honest description of active ownership at A$400 billion: past a certain size, a buyer stops taking prices and starts setting them, with the governance load, the disclosure and the conspicuous exit that follow.
The second door moves the problem rather than solving it. Hand the mispricing to a diversified small-company mandate and the constraint climbs a level: the mandate itself must now be large enough to matter to an A$400 billion allocator, and a manager asked to run A$4 billion in a strategy built for A$1 billion must own more of each float, trade over longer periods, or shrink every active bet until the strategy is the index with extra steps. The constraint has simply moved house, from the security to the mandate, and its new address is where it does the quietest damage, because a mandate dying of scale can report clean numbers for years.
The third door is the one the system is walking through in plain sight. In 2025, for the first time, the funds in NAB’s industry survey, 37 of them holding A$2.69 trillion and more than 80% of the APRA-regulated industry’s assets, reported allocating more than half their portfolios to international assets: 50.9%, up from 47.8% two years earlier and 41% in 2019.⁷ The most cited reason is the diversification and breadth of opportunity available offshore. Australia’s concentration in banks and miners would push any diversified fund offshore. Size helps determine how far that journey runs. Capacity rarely appears in survey answers under its own name; from where we sit, breadth at this size is what it looks like.
Australia is not the first system through this door, and the road beyond it is already mapped. Canada’s large pension plans met the same wall more than a decade ago and answered it in three moves: they internalised, building investment teams in-house; they went direct, into infrastructure, property and private companies, where a giant cheque is an advantage rather than a handicap; and they left home. CPP Investments now manages over C$700 billion and holds the large majority of it outside Canada.⁸ Japan’s GPIF, the largest pool of all, accepted the mirror instead: the overwhelming share of its equities is run passively.⁸ The analogy has limits worth stating. Canada’s plans are defined-benefit, answer to no performance test, and never faced Australia’s member-switching discipline. What carries over is the pattern. Past a certain size, a pool’s remaining choices narrow towards scalable beta, internal direct assets, and export.
Australia is visibly on that road. The number of APRA-regulated funds with more than six members fell from 158 to 81 in the five years to June 2025, and the largest five are increasing their share of contribution flows faster than their share of assets.⁹ Each move is rational, and none of them manufactures differentiated return at the new size; each finds a bigger field in which the old amount of it can hide. What is left is fewer, larger buyers, hiring their own teams, owning more of the world and less of anything small.
Which leaves the last door, the one most of the money takes: own the market, at the market’s weights. The regulator helps hold it open. The annual performance test compares each product’s result to a benchmark portfolio the regulator specifies, and a product that fails twice is closed to new members.¹⁰ The test can end a product over tracking error; it has nothing to say about hugging the index. The largest pools therefore face two pressures leaning the same way: their size makes differentiated positions hard to build, and the test makes the attempt dangerous to be seen making. Whatever the mix of causes, the behaviour is on the record. In the most recent industry survey, most funds reported preferring to stay on benchmark in the US technology sector despite their own valuation concerns.⁷ Doubt the price, hold the weight.
So the compulsory flow settles where it can, in the assets able to absorb it at the weights the index assigns, and the market has noticed. This is the part of the story we would treat as already in the price. Commonwealth Bank finished the 2025 financial year as the largest company on the exchange, reporting record profit and trading at about 30 times earnings, roughly double its own domestic peers and JPMorgan, on a price-to-book ratio more than triple the industry median, a valuation Bloomberg’s analysts described as vying to be the developed world’s most expensive bank.¹¹ There are several stories inside that price. Predictable, size-constrained institutional demand meeting a fixed supply of index weight is surely one of them. Nobody can say how much of the premium is that story rather than the others, and that is partly the point: once structural demand is in a price, it stops being separable from it. The next investor, though, receives only the return available at the price then prevailing. Structural demand can be excellent for the asset and ordinary for its newest buyer.
What we would treat as not yet priced is the other side of the same force. Money concentrating where it can be deployed at scale is money vacating where it cannot. Below the top of the index, a growing national savings pool and shrinking institutional active capital can coexist indefinitely, and the second is a direct consequence of the first.
The best objection to all of this is the record, so give it the floor. Scale has paid. System returns have been strong, 10.1% in the year to June 2025 against long-run averages several points lower,⁹ and size bought the two returns that never decay: fee compression, which compounds for members in every market, and access, to internal capability, to direct assets, to transactions an A$5 billion fund never sees. On that reading the doors are the product working. A national default system’s job is the market’s return delivered at the lowest achievable cost, and scale is the best technology anyone has found for delivering it.
We think that objection is mostly right, and that it concedes the argument rather than defeating it. The system converges on scalable exposure precisely because that is what scale can still buy. The fee saving is real and permanent; it is also capped, a few dozen basis points won once rather than a stream that grows. The unlisted access is real too, and increasingly crowded, because every giant pool on earth has read the same map and is bidding for the same airports. An advantage every buyer of the same size identified in the same decade stops being an advantage; it becomes an auction. The giants keep real judgement, too: asset allocation, implementation and governance remain levers at any size. What narrows is everything beneath them, the security and the strategy. Accept every word of the objection and the conclusion stands. The largest pools are leaving the business of being different in listed markets because their size has priced them out of it. They will cope; the live question is where the capacity they have outgrown ends up.
It ends up downstream, and this is the part that matters in practice for anyone not running A$400 billion. The strategies the mega funds outgrow do not disappear when they leave; the mispriced A$2 billion company from the meeting above is still mispriced after the largest funds conclude they cannot own enough of it to care. Where real mispricings sit below the giants’ reach, capacity constraints transfer them down the size curve, to smaller funds, family offices and specialist managers, for whom an A$200 million position is a result-changing idea rather than five basis points. The same force is reshaping manager selection. With 81 funds where there were 158, an external manager survives on a capability the client cannot internalise, offered at a size that matters to the client without destroying the edge that made it worth hiring. Capacity is part of the product now, as much as the track record is, and the honest managers say so before they are asked.
Both sides of this argument leave public tracks. On one side: the offshore share continuing to climb past half; the fund count continuing to fall; internalisation announcements; substantial-holder notices from the largest funds clustering at the top of the index and thinning out below it; external mandates growing bigger, fewer and cheaper. Each would be the system doing what scale predicts. On the other side, the view weakens if the largest funds keep expanding domestic active allocations while their positions consume no more of each company’s float, if a strong run of new listings widens the shelf faster than the pool grows, or if the offshore share stabilises even as assets compound. That record would say capacity is scaling with the pool after all, and this concern would join the list of limits Australian super was supposed to hit and grew straight through.
A system heading for A$6 trillion will run short of something subtler than assets: strategies that can still matter to it. Every year of compulsory inflow and compounding pushes more of the market’s opportunity set below the size at which the biggest buyers can use it, and their success is the constraint. One per cent of the system is a company. At June’s pace, it is also about a week of the exchange’s entire cash-market turnover.² Origin’s shareholders saw a preview of the end state in a single afternoon, a pool large enough that its vote was the outcome. The closer a buyer comes to being the market, the less room it has left to beat it. Nobody has yet found a way to outperform their own reflection.
General information only. Not personal advice. Past performance is not indicative of future performance. Examples are illustrative. This material is intended for wholesale and professional investors.
Notes
1. Australian Prudential Regulation Authority, Quarterly Superannuation Performance Statistics, December 2025 quarter: total superannuation assets of A$4,485.5 billion at 31 December 2025, of which A$3,181.4 billion was held in APRA-regulated funds; total contributions of A$220.8 billion and benefit payments of A$139.9 billion in the year to December 2025. APRA defines net contribution flows as contributions plus net benefit transfers less benefit payments. The rounded A$80 billion figure used in the text is the simple difference between gross contributions and benefit payments, rather than APRA’s defined net-contribution-flow measure.
2. ASX historical market statistics: total market capitalisation of ASX-quoted entities of A$3,257,972 million at 30 June 2026. ASX Group, Monthly Activity Report, June 2026: total average daily cash-market value traded of approximately A$10.3 billion, across equity, exchange-traded product and interest-rate market transactions.
3. Bloomberg analysis, 2026: Australian superannuation assets projected to reach approximately A$6.0 trillion by 2035. AustralianSuper’s published materials separately state the sector is expected to exceed A$6.1 trillion by 2035. Bloomberg’s baseline measure of the system is narrower than APRA’s total, which includes self-managed funds, so the two levels are not directly comparable.
4. AustralianSuper’s published figures: more than A$410 billion in assets under management at 31 December 2025, with assets forecast by the fund to exceed A$600 billion in 2030. The A$400 billion illustration is deliberately rounded and does not describe an actual AustralianSuper position.
5. Corporations Act 2001 (Cth), s606: a person must not acquire a relevant interest in a listed company’s voting shares that takes their voting power from 20% or below to above 20%, other than through a permitted mechanism such as a takeover bid; substantial-holder disclosure obligations apply above 5%.
6. Origin Energy scheme meeting, 4 December 2023: the scheme of arrangement proposed by the Brookfield and EIG consortium, valuing Origin at about A$20 billion on an enterprise-value basis, failed to achieve the required 75% shareholder approval, with 68.92% of votes cast in favour. AustralianSuper, which had built a holding of about 17%, had stated publicly that the offer was below its estimate of Origin’s long-term value. Sources: Origin Energy’s ASX announcements and contemporaneous Reuters reporting.
7. National Australia Bank, Super Insights Report 2025, the twelfth biennial survey, drawing on 37 funds with A$2.69 trillion under management, more than 80% of APRA-regulated industry assets excluding self-managed funds: allocation to international assets reached 50.9%, crossing half for the first time, up from 47.8% in 2023; NAB’s FX Strategy publication of February 2026 records the rise from 41% in 2019. The 2025 report also found most funds preferred to remain on benchmark in the US technology sector despite valuation concerns, and cited diversification and breadth of offshore opportunity as the leading reason for the international shift.
8. CPP Investments, fiscal 2025 annual report: net assets of approximately C$714 billion, with the large majority invested outside Canada. Government Pension Investment Fund of Japan, annual report for fiscal year 2024: about 90% of GPIF’s equity investment is in passive funds.
9. Australian Prudential Regulation Authority, Annual Superannuation Bulletin 2024–25: the number of APRA-regulated funds with more than six members fell from 158 to 81 over the five years to June 2025; the annual rate of return for the year to June 2025 was 10.1%, against five-year and ten-year averages of 7.8% and 6.5%. APRA has separately noted that the five largest funds are increasing their share of superannuation-guarantee contribution flows faster than their relative share of net assets.
10. Australian Prudential Regulation Authority, Your Future, Your Super performance test: an annual test of products against regulator-specified benchmark portfolios; a product that fails the test in two consecutive years is closed to new members.
11. Reuters, August 2025, citing LSEG data: Commonwealth Bank of Australia reported record full-year cash earnings of A$10.25 billion while its shares traded at about 30 times earnings, roughly double its Australian peers and JPMorgan, and at a price-to-book ratio more than triple the industry median. Bloomberg Intelligence, “CBA Set for Worst Return Potential Among Global Bank Stocks”, 7 March 2025, separately described CBA as vying for the title of the world’s most expensive developed-market bank.
