Backing emerging managers in the era of consolidation: why the small end of the barbell needs a different vehicle, not just a different LP
TLDR: In the first half of 2026, not a single dedicated Southeast Asia private equity fund reached a final close, while three pan-Asia mega-funds absorbed roughly 85% of all the private equity capital raised in Asia Pacific. Read as a verdict on the region, this looks like the end of the small manager. Read correctly, it is a verdict on a particular fund model, and on an allocation arithmetic that was never going to work for the largest LPs. The small end of the barbell is not waiting for pension funds to change their minds. It needs a different buyer, a different source of return, and a different vehicle: one that lets investors back specific companies, with the evidence in front of them, rather than a blind pool.
Zero. That is the number of dedicated SE Asia private equity funds that held a final close in the first six months of 2026, according to DealStreetAsia's Southeast Asia Private Capital Funds: H1 2026 Review. One venture fund made it over the line, at US$8 million. In the same period, EQT's BPEA Fund IX closed at US$15.6 billion, Blackstone Capital Partners Asia III at US$13.1 billion, and Bain Capital Asia Fund VI at US$10.5 billion. Three managers, US$39.2 billion, roughly 85% of everything raised for private equity in the region.
DealStreetAsia's conclusion is blunt: institutional appetite for Asia persists, but "appetite for dedicated Southeast Asia mandates has evaporated."
This is the backdrop to one of the more uncomfortable panels at SuperReturn Asia this month, on backing emerging managers in an era of consolidation. The framing is the barbell: large, dominant platforms at one end, small and nimble managers at the other. The question is how LPs can justify the small end when all the capital is flowing to the large one.
Matsu Partners sits at the small end of that barbell, investing in and operating owner-run SMEs across Southeast Asia and Hong Kong. We do not run a fund. We syndicate each investment to private investors who back us, company by company. What follows is our view of why the capital went where it went, why most of the reasons have little to do with small-manager quality, and what a strategy at our scale owes the investors behind it.
What is actually driving the concentration?
Three things, and none of them is "small managers are worse."
1. Distributions, not returns. The phrase "DPI is the new IRR" became conference shorthand over the last two years for a reason. Distributions from SE Asia dedicated buyout funds have not been great, relying on paper marks. By contrast, the mega-funds could point to realised exits across several Asian markets.
2. Fund dynamics. An LP can deploy at scale to a US$15 billion fund, whilst benefiting from a series of internal scale economies and functionalities (e.g. reporting, value creation teams, etc). A sub-US$200 million fund becomes unattractive given its inability to accommodate meaningful cheque sizes without concentration, whilst lacking large-cap infrastructure. Thus, the small end of the barbell is not a choice they are declining. It is a product they cannot buy.
3. The investment committee has a queue. As DealStreetAsia puts it, regional opportunities "must compete with China, India, Japan, South Korea, Australia and other markets at the investment committee." A pan-Asia fund can rotate between those markets on the LP's behalf. A country or sub-regional fund asks the LP to make that call themselves, and defend it.
Exit markets, concentration limits and committee bandwidth are structural, and none of them tells you anything about whether a smaller manager will make money.
Is scale as safe as it looks?
There is a quiet assumption behind the barbell metaphor that the large end is the safe end. It deserves more scrutiny than it usually gets.
A US$15 billion fund has to deploy US$15 billion (or more like US$30bn incorporating leverage). In a region where exit markets cannot absorb large listings, that is deployment pressure into a narrow set of assets large enough to matter, in competition with two other funds of the same size raised in the same quarter. Style drift follows deployment pressure as reliably as fee drag follows fund size. And an LP who holds all three of this half-year's mega-funds has not diversified; they own the same handful of themes, sourced through the same intermediaries, exiting into the same buyers, three times.
None of this makes the large funds a bad investment. It makes concentration itself a risk that is not really being priced.
Why does the small end still earn its place?
Four reasons, stated plainly, with the honest caveat attached.
Size: Below roughly US$50 million of enterprise value, the giants are still competing - albeit via a PortCo bolt-on strategy. However, this isn't necessarily attractive to owners who particularly want to grow their business to achieve far more consequential value (be it on exit, or surrounding wider consolidation).
Alignment: is real at small scale in a way it is not at large scale. When management fees do not cover the partners' cost of living, the carry is the business plan.
Discipline: is enforced by circumstance. A small fund cannot over-deploy into a hot year because it does not have the capital. A syndicated model cannot over-deploy at all, because every investment has to be underwritten, on its own merits, by the investors who fund it.
Fatter tails: The return distribution for first- and second-time managers has, over multiple cycles and geographies, a fatter right tail than the institutional core. That is the case for the small end.
That comes with a caveat - dispersion. The same distribution has a fatter left tail, and an LP who picks the wrong small manager does not get a mediocre outcome, they get a zero. Anyone selling the small end without acknowledging that is selling something else. This is where an LP must be convinced of the emerging manager's competitive advantage(s) - which itself cannot just rely on asset selection alone.
Can large LPs actually justify the small end, and if not, who can?
For a large institution, a US$75 million commitment to a mega-fund costs one investment memo, one operational due diligence review, one set of references, and one quarterly reporting relationship for a decade. Splitting the same US$75 million across ten emerging managers costs ten of each, for positions that individually cannot move a US$20 billion portfolio and collectively demand more monitoring than the rest of the private markets book combined. Add concentration limits that cap any single ticket at a fraction of a small fund, and you arrive at a simple conclusion: for the largest LPs, the small end of the barbell is uneconomic regardless of how well it performs.
This is not a failure of nerve. It is a cost-per-dollar-deployed problem, and no amount of exhortation about "overcoming risk aversion" will change it, however much AI is said to be simplifying diligence.
What it does change is the answer to who the small end is for. The features that make a US$30–75 million strategy uninvestable for a sovereign wealth fund make it well suited to a different set of allocators: single- and multi-family offices, high-net-worth individuals investing directly, and smaller endowments and foundations.
For these LPs, a US$1–5 million commitment is a meaningful position rather than a rounding error. There is no career to protect from an investment committee, so the decision can be made on the basis of a person and a model rather than a track-record template. And they generally want the thing that large institutions treat as a nuisance: direct ownership in the individual companies, with full visibility, and the right to decide deal by deal. Syndication gives them that as the default rather than as a side letter.
The implication is the important one. The small end of the barbell is not a scaled-down institutional product. It is a different product for a different buyer. The small managers who fail to close are, more often than not, the ones trying to sell it to the wrong one.
What does a manager at this scale owe the LPs who back it?
If the small end is a different product, it should be built like one. Speaking from our own seat, we think a manager at this scale owes its LPs four things, and we will use our own model as the worked example rather than pretend to neutrality.
1. Returns that come from inside the business. At SME scale, you do not reliably control the entry multiple, and you certainly do not control the exit multiple. The one lever that is yours is what happens to revenue and cost while you own the position. This is why we built Matsu around operational value creation rather than financial engineering: applying data science and AI to how an owner-run business acquires customers, prices, collects, and staffs. It is the toolkit that large-cap private equity has deployed for a decade at the top of the market, brought down to the segment where it has never been affordable.
2. Risk that is observable before capital is committed. The two things LPs most fear in an emerging manager are unproven judgment and the absence of a track record. Our answer is to work alongside an owner for up to 90 days, inside the business, demonstrating operational improvement before any equity changes hands. The LP does not have to take our judgment on trust. They can see what moved, and by how much, company by company. It is a track record built per business rather than per fund, and it is underwritable in a way that a fund-III promise is not.
3. An exit path that exists. The DealStreetAsia data is, at root, about a fund model that depended on public listings in markets too shallow to provide them. SME exits do not. They go to trade buyers, corporate strategics and regional consolidators, which increasingly means the same large-cap funds running buy-and-build programmes behind a dominant platform. The same report counts 81 acquisitions across the region in the first seven months of 2026 even as listings stalled. Healthcare, education and business services in Hong Kong, Malaysia, the Philippines, Singapore, Thailand and Vietnam are fragmented, under-digitised, demand-resilient, and have a clearly identifiable buyer universe. That is the DPI story, and it does not require a stock exchange to co-operate.
4. A structure that fits the buyer. If the natural investor is a family office or private individual, the structure should reflect it. We syndicate each investment: investors choose which companies they back and at what size, with transparency at the company level rather than through a fund wrapper, and without paying management fees. There is no blind pool, because the investors we work with do not want one. Asking a family office to behave like a pension fund is as much of a category error as asking a pension fund to write a US$5 million cheque.
There are things a manager at our scale cannot offer, and it would be dishonest to leave them out. We cannot absorb US$100 million from a single LP. We cannot provide the comfort of a brand that everyone on the committee has heard of. And we cannot promise that dispersion does not apply to us. What we can offer is the evidence, at the level of individual businesses, on which you can form your own view.
Conclusion: consolidation eras build the next platforms
Every consolidation cycle in private markets has produced the same second act. While capital concentrates at the top, the mid-market platforms of the following decade are founded at the bottom, at a size the giants ignore, funded mostly by family and private capital rather than institutions. The institutions arrive at fund III or IV and pay for the access that the early backers bought cheaply.
Southeast Asia in 2026 is at the beginning of that cycle, not the end of it. DealStreetAsia is right that appetite for dedicated regional mandates has evaporated. But appetite followed the exits, and the exits at SME scale are the ones still happening.
The small end of the barbell is not empty. It is simply not for everyone.