Multi-Boutique Asset Management: The Architecture of Autonomous Capital

Section 1: The Founder's Dilemma and the Upstream Signal

In December 1993, a former Boston Company executive named William J. Nutt set up a company in Massachusetts with a narrow and unglamorous purpose: to solve the succession and transition problems facing the founders and owners of mid-sized investment firms.¹ He called it Affiliated Managers Group. TA Associates put up the initial backing.¹

The problem Nutt had identified was specific, and it had nothing to do with markets. A successful investment boutique founded in the 1960s or 1970s reaches a moment — usually when the founding generation turns sixty — where two things become true at once. First, the founders' wealth is trapped inside an illiquid private partnership that has no obvious buyer. Second, the next generation of portfolio managers, the ones actually generating the returns, own very little of it. Those two facts create a slow-motion crisis. The founders need liquidity. The successors need ownership. And the only buyers with checkbooks large enough — bank holding companies, insurers, and diversified financial conglomerates — wanted to fold the firm into a corporate parent, rebrand the funds, install a group investment committee, and put the portfolio managers on a standard compensation grid.

That last step is where the value went to die. An investment firm is a balance sheet of people. Put those people on a corporate pay scale, and the good ones leave. When they leave, the track record stops compounding, and the mandates follow them out the door. The acquirer had bought a machine and then removed the engine.

Nutt's structure attacked exactly that failure. AMG would buy an equity interest — sometimes a minority, sometimes a majority — but the affiliate would remain a separate legal entity with its own brand, its own investment process, its own hiring, and its own culture.¹ Members of each affiliate's management retained significant ownership in their own firm.² AMG would take a contractual slice of revenue and provide help with distribution, product development, and operations only when asked.² It was, in effect, a way to sell part of the firm without selling the firm.

The market took some convincing. AMG's first genuinely large transaction, in 1997, was a $300 million cash purchase of a 71% position in Tweedy, Browne, the New York value house founded in 1920.¹ Later that year, in November, AMG completed a $202 million initial public offering and listed on the New York Stock Exchange.¹³ Thirty-three years on, that structure supports $942.4 billion in client assets and a business whose economics look very little like the one Nutt started.⁴

Why we looked here

The belief that made this industry worth investigating can be stated as one proposition, and it is falsifiable: the asset management industry is splitting permanently into commoditised passive beta at one end and high-fee specialist capability at the other, and centralised corporate managers cannot hold elite investment talent through generational transitions — which means the durable economics migrate to structures that buy the cash flow while leaving control with the people who generate it.

Three independent lines of evidence pointed the same way.

The first is allocation data. Willis Towers Watson's Thinking Ahead Institute has tracked the world's seven largest pension markets for decades. In 2000, roughly 7% of those assets sat in private markets and other alternatives; by 2020 it was 26%, and the money came almost entirely out of equities, whose share fell from 60% to 43% while bonds barely moved.⁵ Global pension assets reached a record $58.5 trillion in the Institute's 2025 study.⁶⁷ PwC's asset and wealth management research, which surveyed 264 asset managers and 257 institutional investors across 28 countries, projected alternatives compounding at 6.7% a year to $27.6 trillion by 2028, against 5.9% for the industry as a whole.⁸ Treat the second number as a forecast rather than an observation — but the direction is corroborated by the twenty-year allocation record, which is not a forecast at all.

The second line is talent behaviour, and here the evidence is weaker than the story usually told about it. There is no public dataset that reliably counts portfolio managers leaving conglomerates at a particular AUM threshold; claims of that precision should be treated as folklore. What is observable is the corporate wreckage. Franklin Resources spent the June 2026 quarter reporting $18 billion of long-term net inflows across the group — a good result — of which $1 billion of outflows came from a single acquired affiliate, Western Asset Management; excluding it, the figure was $19 billion.⁹ One damaged franchise inside a $1.79 trillion house was enough to swing the headline.⁹ That is the mechanism, visible in a filing.

The third line is where other people's capital went. GP stakes — buying a permanent minority interest, typically 10% to 30%, in an alternative manager's general partnership and taking a pro-rata share of management fees, balance-sheet returns and carried interest — began as a curiosity.¹⁰ Goldman Sachs raised about $1 billion for the first Petershill fund in 2007; Dyal Capital Partners spun out of Neuberger Berman in 2011 with a $1.3 billion debut.¹¹ By the fourth quarter of 2021 more than $20 billion was raised across GP stakes funds in a single quarter, Dyal had merged with Owl Rock to become Blue Owl via a listed vehicle, and Petershill had floated in London.¹¹ The trade was still live in March 2026, when Atlas Holdings took GP stakes backing from Blackstone and Blue Owl simultaneously.¹² When several sophisticated pools of capital independently converge on buying non-controlling economics in investment firms, that is a signal about where the returns are.

How it transmits

The belief reaches this industry's cash flows through three channels. Capital moving from liquid long-only equities into private markets and liquid alternatives raises the fee earned per dollar of assets, so a platform's revenue can grow even when its asset base does not. Founder succession creates a steady supply of sellers who will accept a partner but not an owner, which sets the price of entry. And distribution has consolidated into a smaller number of institutional and wealth gatekeepers whose operational due diligence a $3 billion boutique cannot pass alone, which makes platform membership worth paying for.

The same belief implicates three sibling industries this piece will not chase: dedicated GP-stakes vehicles, wealth-management RIA aggregators, and the turnkey platforms that sit between advisers and managers.

The theme, stated so it can fail

It helps to separate four things that get blurred together.

The structural force is the reallocation of institutional and wealth capital from liquid, benchmark-constrained active management into private markets and liquid alternatives, plus the demographic fact that a generation of boutiques founded between 1970 and 2005 needs an ownership solution. The investable theme is narrower: platforms that buy permanent economics in independent managers without taking control. The industry boundary excludes fully integrated managers like BlackRock, whose model is a single brand and a single operating platform, and excludes wealth aggregators, which buy distribution rather than manufacturing. The security expressions are a small handful of listed companies, of which one — AMG — is close to a pure play and the rest are partial.

The causal chain is testable at every link. Allocators shift money into higher-fee, longer-locked strategies; independent managers of those strategies need distribution and succession capital but refuse integration; platforms supply both in exchange for a contractual slice of revenue; that slice arrives with almost no cost attached, converts to free cash flow, and is recycled into more stakes or into the platform's own shares. Break any link and the theme fails. If allocators stop shifting, link one goes. If GP-stakes funds outbid on price, link three stops earning a return. If the platform overpays or stops repurchasing, link five stops reaching shareholders.

The theme has a clock. Private markets fundraising cycles run eighteen to thirty months, consultant manager reviews run quarterly, and evergreen wealth subscriptions report monthly. That means the theme's health is observable at a quarterly cadence, and a genuine break should show up within four quarters rather than four years.

Adoption is harder to measure than most write-ups admit. There is no reliable published denominator for "independent boutiques of $10–50 billion partnered with a platform versus standing alone," because most of the population is private and does not disclose ownership. Any penetration rate quoted to a decimal point is invented. The usable proxies are three: the pace of GP-stakes and minority-stake transactions, the growth of platform-affiliated assets, and the share of wealth-channel model portfolios allocated to private markets. All three are directionally observable; none is precise.

A note on the reader's problem, because it governs everything that follows. This is written for general institutional public-equity diligence over a three-to-five-year horizon, across global listed expressions, with no position sizing implied. A benchmark-relative long-only reader and an absolute-return long/short reader face different versions of this question — the first is asking whether to hold an underweight in a sector that has just run, the second whether the dispersion between multi-boutique and pure-play alternative managers has gone too far. That matters, because the central risk in a theme like this one is subtle: you can be right about the industry and lose money anyway, by owning the wrong layer, the wrong capital structure, or the right company at the wrong price. As we will see, by the summer of 2026 the price question had become the entire question.

↑ Back to Top

Section 2: The Structural Disconnect: Why Monolithic Asset Managers Lose Their Stars

Start with something that actually happened, in the quarter just reported.

In the three months to June 2026, Artisan Partners lost roughly $5.7 billion of net assets from its Value Equity strategy after a single US sub-advisory mandate was terminated, and began an orderly wind-down of that strategy.¹³ Artisan is not a bureaucratic conglomerate. It is a Milwaukee firm built explicitly around autonomous investment teams, each with its own leadership and process, sharing one distribution and compliance backbone. It has, by the standards of this industry, near-model governance.

And it still lost the mandate. One client decision removed an entire strategy.

Hold that against the Franklin Resources number from a moment ago — one impaired affiliate turning a $19 billion quarter into an $18 billion quarter — and you have the industry's two failure modes in one frame.⁹ The conglomerate's problem is that it damages franchises it owns. The boutique's problem is that a single client can end one. Neither structure escapes the underlying fact: in asset management, the product is a promise made by identifiable people, and both the people and the promise are portable.

The mechanics of corporate absorption

Why does integration destroy the thing it bought? The sequence is well documented across three decades of asset management M&A and it runs roughly like this. The acquirer consolidates the middle and back office to fund the purchase price. It migrates the boutique onto the parent's trading, compliance and reporting stack, which means the investment team's operational requests now queue behind the parent's priorities. It puts the funds under the parent's brand, on the theory that brand is an asset — which transfers goodwill from the people to the platform. And it replaces founder equity with corporate stock or capped cash bonuses.

Each step is individually defensible. Together they change the payoff facing a thirty-eight-year-old portfolio manager with a top-quartile five-year record. Before, growing the firm made them richer. After, growing the firm makes the parent's shareholders richer and makes the manager's bonus a negotiation with a compensation committee that also has to pay for a failed technology migration and a shrinking legacy mutual fund range.

This is a principal-agent problem, not a culture problem, and the distinction matters because culture is unfixable by contract while incentives are not. Alpha generation depends on concentrated conviction, fast decisions, and a willingness to look wrong for two years. Every one of those behaviours is career-risky inside a large organisation. The rational response to a capped upside and an uncapped career downside is to hug the benchmark. Benchmark-hugging with an active fee is the single most reliably fatal position in modern asset management, because the passive alternative charges nearly nothing.

What the multi-boutique structure actually fixes

The multi-boutique answer is to leave the operating company intact and buy the cash flow instead. Under AMG's structure — which the company describes as a structured partnership interest — the affiliate stays a stand-alone firm. It allocates a specified percentage of its revenue to AMG and to affiliate management; the remainder funds operating expenses and additional distributions to affiliate management.² For most affiliates, AMG shares in revenue without regard to expenses.² Investment decisions, hiring, research and firm identity stay with the partners. AMG supplies help with strategy, marketing, product development and operations when asked.²

Think of it as a royalty on a mine rather than ownership of the mine. The royalty holder gets a defined slice of what comes out of the ground and does not pay for the drilling. The analogy has a real limit, and it is the limit that matters most for this business: a mine's ore body is fixed and known, while an investment firm's revenue depends on people who can walk out. A royalty on a declining revenue base squeezes the operating pool that pays those people, which is precisely when they walk. We will return to that failure mode, because it is the structural risk that the bull case tends to leave out.

What the structure does buy is a genuinely different negotiating position at the point of sale. A founder choosing between a strategic acquirer and AMG is not comparing two prices for the same thing. One buyer takes control; the other takes economics. Hamilton Helmer would call this counter-positioning: AMG's model is unattractive for an incumbent conglomerate to copy, because a conglomerate's entire investment case rests on the cost synergies that autonomy forbids. Invesco cannot promise a boutique full independence without abandoning the logic that justified buying it.

The moat test

Now stress the claim, because the industry likes to state it as a law.

The popular version says autonomy wins — that independent boutiques outperform and that platform-owned teams decay. Test it against 2026. GQG Partners is about as pure a case as exists: founder-led, independent, listed in Australia, with no corporate parent telling it what to own. It reported US$163.3 billion of funds under management at the end of May 2026, down from US$166.9 billion a month earlier on US$1.9 billion of net outflows and US$1.7 billion of negative performance, and it had run about US$15 billion of outflows over the first half of the year across six consecutive monthly declines.¹⁴¹⁵ Autonomy did not protect it. Performance did not hold, and clients left.

The honest conclusion is narrower and more useful than the slogan. Autonomy protects incentives; it does not manufacture returns. What the multi-boutique structure prevents is the self-inflicted destruction of a good franchise. It does nothing about the ordinary risk that a strategy stops working. An investor buying this theme is buying a better survival rate on acquired assets, not immunity from redemption.

There is a third case worth holding alongside those two, because it complicates the story in a useful direction. T. Rowe Price never solved the succession problem by buying anyone; it grew its own people, kept one brand, and stayed centralised. On the theme's logic it should be the most vulnerable firm in the peer group. It managed $1.89 trillion at 30 June 2026 — more than any other company discussed here except Invesco — and its June net inflow of $0.8 billion, helped by one large sub-advised equity win, still sat inside a quarter of $6.5 billion of net outflows.¹⁶ So the centralised model is losing assets, but slowly, from an enormous base, and with the retirement-plan relationships that make its long-only franchise stickier than a sub-advisory mandate. Structure explains the direction of travel. It does not explain the speed, and a thesis that predicts collapse at monolithic managers has been waiting a long time.

The moat question follows directly. Large managers claim two defences: brand and distribution scale. Both are real and both are conditional. Brand in asset management is a shorthand for trust in a process, and it transfers to whoever runs the process — which is why Artisan's affected clients moved their money, not their loyalty. Distribution scale is genuinely valuable, because gatekeeper shelf space is finite. But shelf space with nothing worth selling on it is a fixed cost, and that is the position an integrated acquirer ends up in when the team it bought has left. The multi-boutique answer inverts the order: acquire the capability first and rent the shelf space to it. AMG's affiliate roster spans long-only equity specialists and alternatives firms alike — AQR, Capula, Garda, Systematica and Winton in liquid alternatives, Pantheon, Arra, EIG and Qualitas Energy in private markets — and the platform's role for each is the same: open the institutional and wealth doors, then get out of the way.¹⁷

Which raises the question that separates the platforms from one another: if you cannot control the investment process, what exactly are you selling, and how do you get paid for it?

↑ Back to Top

Section 3: The Architecture of Autonomy: Revenue-Sharing vs. Operating Roll-Ups

There are four live answers to that question in the listed market, and they differ in ways that show up directly in the income statement.

The revenue-share partner. AMG's contract takes a defined percentage of an affiliate's top-line revenue before expenses.² The consequence is subtle and important. If an affiliate hires ten analysts, opens a Singapore office, or overpays for space, none of that touches AMG's share — the cost comes out of the pool the partners control. AMG's economics are therefore insulated from affiliate cost inflation in a way that an equity owner's are not. The reverse is also true and less discussed: if affiliate revenue falls, the fixed obligations sitting inside the partners' pool do not fall with it, so the squeeze lands entirely on partner compensation. That is when contracts get renegotiated and stakes get impaired.

The integrated roll-up. Victory Capital buys franchises outright and moves everything that is not investment management onto one platform. Its scale step-change came from the Amundi US transaction, which brought back the Pioneer Investments brand as an investment franchise and lifted assets under management from roughly $170 billion to about $290 billion.¹⁸ By 30 June 2026 Victory reported $342.4 billion of AUM and $346.1 billion of total client assets, with $4.2 billion of long-term net inflows in the second quarter.¹⁹ For the twelve months to March 2026 it generated $1,474.9 million of revenue and $644.5 million of EBITDA — a 43.7% margin on consolidated revenue — with $66.7 million of interest expense, which is the honest price of a debt-funded acquisition model. Victory's second-quarter financial results were scheduled for release on 5 August 2026, after this article's cutoff, so the margin picture here is the March-quarter trailing view.

Note that Victory's partner in that transaction now reports on it from the other side: Amundi disclosed that Victory contributed €35 million to its second-quarter 2026 results through synergies and that Victory strategies drew €2.9 billion of inflows in Asia and Europe.²⁰ That is unusually direct evidence that a roll-up's distribution claim is real rather than aspirational.

The autonomous-team platform. Artisan Partners sits in between. Teams are independent; the corporate entity is one legal and distribution structure. Its June 2026 assets were $183.4 billion, split $93.5 billion in Artisan Funds and Artisan Global Funds and $89.9 billion in separate accounts and other vehicles.¹³ Second-quarter revenue was $307.9 million with GAAP earnings of $1.11 a share, adjusted earnings of $0.94, and a quarterly dividend of $0.80, up 10% year over year.²¹ That dividend is the tell. Artisan is run as a payout vehicle, not a compounder.

The minority incubator. Pinnacle Investment Management in Sydney takes 20%-to-49%-type stakes in early-stage teams and supplies seed capital, distribution, and the responsible-entity and back-office plumbing that a start-up manager cannot build. It closed December 2025 with A$202.5 billion of affiliate funds under management, up 13% from A$179.4 billion six months earlier, and by March 2026 counted 18 affiliates and roughly A$208 billion.²² First-half net inflows were a record A$17.2 billion against A$6.7 billion in the prior comparable period, and aggregate affiliate funds-management revenue on a 100% basis rose 24% to A$546.6 million.²² Two cautions before anyone treats that as the best growth in the peer group: the figures are Australian dollars, the revenue is measured across affiliates Pinnacle only partly owns, and Pinnacle's own first-half profit fell.²² Growth in affiliate FUM and growth in the parent's earnings are different things.

A smaller Australian cousin, Pacific Current Group, shows the model's other end. Its continuing boutiques held A$26.4 billion at 30 June 2026, up 1.8% in the quarter, while total FUM fell 6.4% because Pacific Current exited its Aether boutique in June, taking A$2.3 billion of assets with it.²³ At that scale, one disposal moves the whole number — a reminder that minority-stake portfolios are lumpy, and that "boutique platform" spans a very wide range of economic substance.

Reading the margins honestly

The dossier version of this comparison says AMG runs a 51.2% adjusted EBITDA margin against Victory's 48.5%, and concludes that revenue-sharing is structurally superior. That comparison does not survive contact with the accounting.

AMG reported $316.0 million of adjusted EBITDA on $640.7 million of consolidated revenue in the second quarter of 2026.⁴ That ratio is 49%, but it is not a margin in any meaningful sense, because adjusted EBITDA includes AMG's share of earnings from affiliates whose revenue never appears on AMG's income statement at all. The denominator and numerator describe different businesses. Victory's 43.7% is a real margin — its revenue and its costs are both consolidated.

So the correct statement is narrower: AMG converts a higher proportion of its reported revenue into adjusted EBITDA because much of its economics arrives already net of affiliate costs, and Victory earns a genuine mid-forties operating margin on a fully consolidated cost base. Both are good. Ranking them on a single percentage is a category error, and it is exactly the sort of error that makes a screen look attractive before the diligence starts.

The more revealing structural comparison is what each model does with a bad affiliate. Victory can fire management, merge the strategy, and cut the cost base. AMG can do almost none of that; its remedy is to renegotiate the revenue share or write down the intangible. Control has a value, and it shows up in a downturn.

Where the power actually comes from

It is worth being precise about which competitive advantages here are real and which are narrative, because "we have a differentiated model" is the most over-claimed sentence in asset management.

Run the five forces over the multi-boutique platform and most of them are unkind. Buyer power is high and rising: institutional allocators have professionalised fee negotiation and increasingly demand co-investment rights that dilute the manager's economics. Supplier power — the investment teams — is also high, because the scarce input is a person who can leave. Substitutes are brutal: an index fund does 80% of what a long-only affiliate does for a small fraction of the price. Rivalry among platforms bidding for stakes has intensified, as the GP-stakes fundraising record shows.¹¹ Only barriers to entry offer comfort, and even there the barrier is capital and reputation rather than anything structural.

So where does the advantage live? Two places, and neither is scale.

The first is counter-positioning, in the sense Hamilton Helmer uses it: AMG offers something an integrated acquirer cannot match without destroying its own economics. A conglomerate justifies an acquisition by cost synergies, so it cannot credibly promise autonomy. That asymmetry has held for three decades and is the most durable thing in this business. What could erase it: a GP-stakes fund offering the same autonomy with a lower cost of capital and no distribution overhead — which is precisely what Blue Owl and Petershill do.¹¹

The second is switching cost at the affiliate level, and it is contractual rather than emotional. A structured partnership interest is a permanent claim on revenue. Affiliate partners cannot simply buy it back when they decide they would rather keep the money, and the revenue share survives changes in the affiliate's leadership.² That permanence is why the model's cash flows are more predictable than the underlying businesses.

What is not an advantage, despite frequent claims to the contrary: the distribution platform. Fifty AMG professionals working with 500 affiliate specialists is useful, and $125 billion of institutional gross sales since 2009 is evidence it works.²⁴ But nothing about it is proprietary. Any well-capitalised platform can hire a distribution team. It is a service, priced accordingly, and it should not be mistaken for a moat.

Which brings us to the decision that actually made AMG's last decade: what it chose to buy.

↑ Back to Top

Section 4: The Great Asset Pivot: From Liquid Active Equity to Private Markets and Alts

In 2020, alternatives — private markets plus liquid alternative strategies — produced roughly 35% of AMG's EBITDA.²⁴ By the first quarter of 2026 they produced 58%, and management put the figure above 60% by the second quarter, up from about 50% eighteen months earlier.²⁴²⁵ Over the same stretch, adjusted EBITDA went from $795 million in 2020 to $1,077 million in 2025, a 6% compound rate, while economic earnings per share went from $13.30 to $26.05, a 14% compound rate.²⁴ The gap between those two numbers is the buyback, and we will come to it.

That is a business being rebuilt underneath a stable name.

Exhibit 1 — AMG earnings and asset mix, 2020 vs. 2026 Definition: alternatives = private markets plus liquid alternative strategies. Earnings share measured as contribution to EBITDA; asset figures are period-end AUM in US$ billions. Geography: global. Evidence status: company-disclosed.

Measure	2020	Q1 2026 (31 Mar)	Q2 2026 (30 Jun)
Alternatives share of EBITDA	~35%	58%	>60%
— of which private markets (LTM EBITDA)	n/d	39%	n/d
— of which liquid alternatives (LTM EBITDA)	n/d	19%	n/d
Private markets AUM ($B)	n/d	148	n/d
Liquid alternatives AUM ($B)	n/d	261	n/d
Differentiated long-only AUM ($B)	n/d	474	n/d
Total AUM ($B)	n/d	882	942

Sources: AMG first-quarter 2026 disclosure and second-quarter 2026 results and earnings call.²⁴⁴²⁵ "n/d" = not disclosed in the cited materials.

Read that aloud and the shape is clear. Alternatives are 46% of AMG's assets but nearly two-thirds of its earnings. Long-only strategies are the majority of the asset base and a minority of the profit. That single divergence is the entire investment case, and it also tells you the vulnerability: a business whose earnings are concentrated in a minority of its assets is more exposed to what happens in that minority than the asset mix suggests. Note also that the dossier's widely circulated "65% alternatives" figure is not what the company disclosed; the disclosed readings are 58% and "more than 60%."²⁴²⁵ The difference is small, but on a metric being used as a kill threshold it is worth getting right.

Why the fee arithmetic works

The reason the earnings mix diverges from the asset mix is price. A long-only institutional equity mandate competes against an index fund charging a handful of basis points, and its fee has been compressed for fifteen years. A private infrastructure fund charges a management fee on committed capital for seven to ten years and takes a share of profits above a hurdle. The second contract is worth several times the first per dollar, and — critically — the capital cannot leave when performance disappoints.

That last point is the one investors under-weight. The value of a private markets affiliate is not only its higher fee; it is that its revenue base is contractually locked. A long-only affiliate can lose a third of its assets in a quarter, as Artisan's Value Equity strategy did.¹³ A closed-end infrastructure fund cannot.

AMG does not publish a blended realised fee rate in basis points, and any figure quoted as such should be treated as an estimate rather than a disclosure. Dividing consolidated revenue by average AUM produces a meaningless number, because the revenue line excludes the affiliates accounted for under the equity method. This is a genuine disclosure gap, and it is the reason the market has historically struggled to value the company.

What AMG actually bought

The pivot was executed through a series of minority partnerships rather than one transformative deal. Pantheon brought private equity, infrastructure and secondaries. EIG brought energy and energy-infrastructure investing; AMG acquired a minority stake in 2014 with EIG management retaining majority ownership.²⁶ Systematica added quantitative liquid alternatives. Comvest added middle-market direct lending. In May 2025, AMG agreed to acquire a minority equity interest in Qualitas Energy, a renewable and energy-transition infrastructure platform, again leaving the management team with majority ownership and day-to-day control.²⁷

The pattern is deliberate: minority stakes, management keeps control, no integration. It also means AMG competes for these assets against the GP-stakes funds described earlier — Blue Owl, Petershill, Blackstone's strategic capital arm — which is why entry multiples matter as much as entry quality.

The flow evidence in 2026 shows where the demand actually sat. In the first quarter, AMG's liquid alternatives affiliates took $25 billion of net inflows, with AQR, Capula, Garda, Systematica and Winton all contributing, while private markets affiliates raised about $4 billion, led by Pantheon secondaries and infrastructure fundraises at Arra, EIG and Qualitas Energy.¹⁷ Total net client cash flow was $22.5 billion for the quarter against a $0.4 billion outflow a year earlier.²⁴ In the second quarter, alternatives inflows reached $29 billion while total net flows were $12.9 billion — meaning long-only strategies bled roughly $16 billion in three months.⁴²⁵ Over the trailing twelve months alternatives took in about $100 billion.²⁵

That is the transformation stated in its most honest form: a very large alternatives franchise growing fast, attached to a shrinking long-only franchise, inside one holding company.

The roads not taken

The traditional houses answered the same passive threat differently, and the results are instructive.

Franklin Resources bought its way into scale and into alternatives, and now manages $1.79 trillion.⁹ Its problem is that acquired breadth brought acquired fragility, which is what the Western Asset drag illustrates.⁹

T. Rowe Price largely declined to buy scale at all and remains the cleanest expression of high-quality traditional active management. It managed $1.89 trillion at 30 June 2026, took $0.8 billion of net inflows in June alone, and still ran $6.5 billion of net outflows for the quarter.¹⁶ The June figure included a large sub-advised equity inflow — a reminder that a single mandate can flatter a month.¹⁶

Invesco did something the alternatives narrative tends to ignore: it won by leaning into passive. Second-quarter 2026 net inflows were a record $45.1 billion on $2.5 trillion of assets, up 14.4% year on year, with ETFs and index strategies at a record $753.5 billion contributing $30.1 billion, the QQQ product $13.8 billion, and the China joint venture $6.9 billion on record assets of $163.2 billion.²⁸ Private markets contributed $1.9 billion.²⁸ Operating margin expanded about 470 basis points.²⁸

Sit with that comparison. In the same quarter, AMG took $12.9 billion of net flows led by high-fee alternatives, and Invesco took $45.1 billion led by low-fee index products — and Invesco's margins went up.⁴²⁸ The bifurcation thesis says both ends of the barbell win. That quarter, both ends did. What it does not say, and what the data supports, is that the passive end is winning on a far larger volume base.

Meanwhile the specialists AMG is increasingly compared with had a harder year. StepStone closed fiscal 2026 with $233 billion of assets, up 23%, on $38 billion of gross capital formation, with its private wealth platform approaching $18 billion.²⁹ Hamilton Lane crossed $1.0 trillion of assets under management and supervision as of 31 December 2025, though only $146.1 billion of that was discretionary — the rest is advisory, which carries a fraction of the fee.³⁰ Both are legitimate private-markets businesses. Both, as we will see, were treated brutally by the stock market in 2026.

At the top of that comparison sit the two firms every alternatives conversation eventually reaches. Blackstone manages roughly $1.3 trillion and is the industry's scale benchmark across real estate, credit and private equity; KKR manages roughly $744 billion across a similar spread plus a large insurance balance sheet.³¹ They matter to this story in two ways. They set the price of the earnings stream AMG's alternatives affiliates produce, which is why the valuation gap between them and AMG is the theme's central financial claim. And they compete, through their strategic capital arms, for the same minority stakes AMG buys — Blackstone was one of the two backers of Atlas Holdings in March 2026.¹² For AMG shareholders, these firms are simultaneously the valuation benchmark and the bidder on the other side of the table.

The capital cycle underneath the adoption curve

Adoption curves and capital cycles rarely peak together, and here they have visibly diverged.

On the adoption side, the evidence looks like deployment: allocators are still increasing private markets weightings, wealth-channel vehicles are scaling from a small base, and the flow data supports it. On the capital side, the evidence looks late. GP-stakes capital raised more than $20 billion in a single quarter at the 2021 peak; the strategy has since institutionalised, with Investcorp raising more than $700 million for a first fund and Blue Owl approaching roughly $2.5 billion for a continuation vehicle.¹¹ That is a large amount of dedicated capital chasing a fixed and slow-growing population of high-quality mid-sized managers.

The arithmetic of that is unforgiving. When several well-funded buyers pursue the same scarce asset, the price rises and the return on newly deployed capital falls, regardless of how well the underlying assets perform. This is the classic pattern in which end-market adoption exceeds expectations while the capital deployed into serving it earns progressively less. It is also why the most informative capital-allocation signal from a platform in 2026 was not what it bought but what it did not: AMG repurchased $375 million of its own stock in the first half against roughly $600 million planned for the year, while adding one minority stake announced the prior May.⁴²⁵²⁷ A buyer that finds its own shares cheaper than the assets it hunts is telling you something about the acquisition market.

↑ Back to Top

Section 5: Value Chain Mechanics and the Profit-Pool Waterfall

Follow one dollar and the industry's power structure becomes legible.

A pension fund or sovereign wealth fund decides to allocate to private infrastructure. That is the top of the chain, and it is where the money and the ultimate bargaining power originate. Asset owners have spent two decades shifting exactly this way — the move from 7% to 26% in alternatives across the largest pension markets is the aggregate of millions of such decisions.⁵ They increasingly demand co-investment rights, fee transparency, and direct access to the investment team rather than a relationship manager.

To reach a manager, that dollar usually passes a gatekeeper. For institutional money it is an investment consultant or the fund's own manager-research staff; for wealth money it is a wirehouse platform, a private bank, or an RIA's investment committee. This layer does not manage money. It decides which managers are permitted to be considered, and it screens on operational infrastructure, compliance depth, business continuity, and firm stability as much as on returns. A three-person boutique with a superb record can fail that screen on the operations question alone.

This is where a multi-boutique platform earns its keep, and AMG quantifies it. The company has generated roughly $125 billion of institutional gross sales since 2009 and roughly $150 billion of wealth gross sales since 2005, supported by more than 50 AMG distribution professionals working alongside more than 500 sales and marketing specialists employed by the affiliates themselves.²⁴ That ratio — 50 against 500 — is the model in miniature. AMG is not the sales force. It is the layer that opens doors the affiliates then walk through, and it is a confirmed, disclosed relationship running from AMG to affiliates including Pantheon and EIG.²⁴¹⁷

Beneath the manager sits the operational layer: custody, fund administration, trade execution, clearing and reporting. Every firm in this industry consumes those services, and the economics are unambiguous — this layer prices in basis points, competes on cost, and has no leverage over a manager that can move its business. Named vendor relationships for specific multi-boutique platforms are not disclosed in the sources reviewed here, so this should be read as a structural description of the layer rather than a claim about any particular contract.

Where the money stops

Now the waterfall. Gross fees are collected by the affiliate. Under AMG's structure, a specified percentage of that revenue is allocated to AMG and to affiliate management, and the remainder funds the affiliate's operating expenses and further distributions to its own people.² The affiliate's pool pays the analysts, the office, the technology and the bonuses. AMG's share arrives before any of that.

What reaches the holding company is close to free cash flow, because the holding company has almost no cost structure of its own. There is no manufacturing, no inventory, no working capital cycle and no meaningful capital expenditure. AMG entered the second quarter of 2026 with roughly $2 billion of available capital, a $1.25 billion credit facility maturing in 2029, credit ratings of A3 from Moody's and BBB+ from S&P, and an average debt duration of about eighteen years.²⁴ That last detail is worth pausing on: eighteen-year average duration on the debt of a business whose revenue can reprice in a quarter is a deliberate mismatch in the company's favour.

From there the cash goes two places: new affiliate stakes, and its own shares. In the second quarter of 2026 AMG repurchased $189 million of stock, $375 million in the first half, against a quarterly dividend of one cent a share.⁴ The dividend is a rounding error by design.

Where the power sits

Three nodes hold real bargaining power, and none of them is the holding company.

The first is a genuinely scarce investment team. If a manager is one of a handful of credible providers of a capability an allocator has decided it needs, that manager sets terms — with clients, and with any platform hoping to buy a stake.

The second is the gatekeeper. Shelf space is finite and access to it is granted, not bought.

The third, newly, is liquidity itself. In February 2026, Blue Owl restricted investor withdrawals from a retail-focused fund; asset manager shares fell across the sector.³² The underlying vehicle had seen redemptions running roughly 20% above the prior year through 2025.³² The lesson generalises: in the evergreen structures now being sold to wealth channels, the manager's ability to control redemptions is the product's risk, and when one manager uses that control, every manager's product is repriced. That is a propagation mechanism with no analogue in the closed-end institutional world, and it is new.

What breaks

Chokepoints in this chain are not physical. They are contractual and reputational, and they propagate fast.

If a flagship affiliate loses its founding partners, institutional consultants place the strategy under review, wealth platforms freeze new allocations, and redemptions begin — typically within two to four quarters, because manager-research committees meet quarterly. The holding company cannot intervene, because intervening is precisely what it contracted away.

If private markets valuations fall, three things happen at once: performance fees stop crystallising, the fundraising cycle lengthens, and the marks on the platform's own affiliate stakes come under audit pressure. Performance fees are already the lumpiest line — AMG earned $49 million of them in the first quarter of 2026, up $29 million year on year, on $240 billion of performance-fee-eligible assets.²⁴ That is a swing factor of real size on a quarterly EBITDA base of roughly $316 million.⁴

And if the operating allocation at a shrinking affiliate stops covering competitive compensation, the partners renegotiate or leave. This is the failure mode the royalty analogy conceals, and it is the one to watch in a long-only affiliate that has lost half its assets.

Who supplies whom, stated plainly

Because this chain has no physical goods moving through it, it is worth naming the directional relationships explicitly.

Boutique founders and partners supply the platform with the thing it cannot make: an investment capability with a track record and a client list. What they receive is liquidity for accumulated ownership, capital for growth, and a succession mechanism. The evidence for this flow is the transaction record itself — AMG's minority purchases at EIG in 2014 and Qualitas Energy in 2025 both explicitly left management with majority ownership and operational control, which is the consideration the founder is buying.²⁶²⁷ Leverage in that negotiation sits with the founder when the capability is scarce and with the buyer when it is not, which is why prices for private credit and infrastructure stakes have risen while prices for long-only equity stakes have not.

The platform supplies its affiliates with institutional and wealth distribution, product structuring, seed capital and operational support delivered on request rather than by mandate.² Affiliates need it because gatekeepers screen on operational depth that a small firm cannot demonstrate alone. Alternatives exist — an affiliate could hire its own distribution team — but doing so consumes the operating allocation that pays its investors, which is precisely the trade the platform relationship is designed to avoid.

Gatekeepers supply the platform and its affiliates with access to end clients, and they are the layer with the most one-sided leverage in the chain: they choose from many managers, and no single manager is essential to them. The clearest public evidence of how much that access is worth is the volume that passes through it — roughly $150 billion of wealth-channel gross sales for AMG affiliates since 2005.²⁴

The custody, administration and reporting layer supplies everyone with the plumbing and captures very little of the economics, because its services are substitutable and priced in basis points. Specific vendor arrangements for the companies discussed here are not disclosed in the sources reviewed, so this remains a structural characterisation rather than a claim about particular contracts.

And at the bottom, the platform supplies its own shareholders — with dividends in Artisan's case, and overwhelmingly with retired shares in AMG's.²¹⁴

↑ Back to Top

Section 6: Competitive Field & Parameter Leadership: Scale, Margins, and Capital Allocation

There is no single leader in this industry, and any table that names one is measuring one thing and implying another. Break it into parameters and the picture sharpens considerably.

Exhibit 2 — Operating scorecard, mid-2026 Definitions: AUM as reported by each company at the stated date; flows are net client cash flows for the stated period; currency as reported. Evidence status: company disclosure except where noted.

Company	AUM (date)	Net flows, period	Notable composition
AMG	$942.4B (30 Jun 26)	+$12.9B Q2; +$35.5B H1	Alternatives +$29B Q2, ~$100B LTM425
Victory Capital	$342.4B AUM / $346.1B client assets (30 Jun 26)	+$4.2B long-term Q2	Pioneer franchise from Amundi US1918
Artisan Partners	$183.4B (30 Jun 26)	−$5.7B from one Value Equity mandate	Strategy in orderly wind-down13
Federated Hermes	$911.6B (30 Jun 26)	n/d in cited release	Money market $676.9B; MMF share 6.7%33
T. Rowe Price	$1.89T (30 Jun 26)	−$6.5B Q2	+$0.8B in June incl. large sub-advised win16
Franklin Resources	$1.79T (30 Jun 26)	+$18B long-term Q2	+$19B excluding Western Asset9
Invesco	$2.5T (30 Jun 26)	+$45.1B Q2 (record)	ETFs/index $753.5B, +$30.1B28
StepStone	$233B (FY26 close)	$38B gross capital formation FY26	Private wealth ~$18B29
Hamilton Lane	$1.0T AUM+AUS (31 Dec 25)	n/d in cited release	Only $146.1B discretionary30
Pinnacle (affiliates)	A$202.5B (31 Dec 25); ~A$208B (Mar 26)	+A$17.2B H1 FY26 (record)	18 affiliates; parent profit fell22
GQG Partners	US$163.3B (31 May 26)	~−US$15B H1 2026	Six consecutive monthly outflows1415

Read across the flow column and the industry's actual 2026 looks nothing like a simple active-versus-passive story. The largest single flow number belongs to Invesco, and it is almost entirely index product. The highest-fee flows belong to AMG's alternatives. The worst flows belong to a fully independent, founder-led equity boutique. Scale, model and outcome are not correlated in the way the theme predicts.

Now the parameter claims, each with a rival and a date.

Alternatives assets held through a non-controlling platform structure. AMG leads, with $409 billion across private markets and liquid alternatives at 31 March 2026.²⁴ The closest listed comparison is Franklin Resources, whose alternatives franchises are substantial but which does not disclose a comparably defined figure in the sources reviewed here, so the ranking is contested on disclosure grounds rather than settled. What is not contested is composition: AMG's alternatives sit in separately branded firms whose management retained majority ownership, as at EIG and Qualitas Energy.²⁶²⁷ Why it leads: fifteen years of buying minority positions in firms that would not have sold control at any price, using a contract form no integrated acquirer can match without undermining its own synergy case. What could erase it: GP-stakes funds bidding the same assets with cheaper capital and no expectation of distribution synergies.

Cash conversion into share count reduction. AMG leads decisively. Adjusted diluted shares fell from 43 million at 31 March 2021 to 27 million at 31 March 2026, a 37% reduction, with roughly $3 billion of capital returned over the period.²⁴ Buybacks were $375 million in the first half of 2026 and management indicated roughly $600 million for the full year.⁴²⁵ The closest rival on capital-return discipline is Artisan, but it does the opposite thing: it pays out, with a quarterly dividend of $0.80, up 10% year over year.²¹ Neither is wrong. AMG's earnings are less predictable and its stock has traded at a low multiple, which favours repurchase; Artisan's are steadier and its shareholder base wants income. Why AMG leads: an asset-light structure that converts nearly all economic income into distributable cash, plus a decade in which its own shares were the cheapest asset available to it.

Absolute net flows. Invesco leads, with a record $45.1 billion in the June 2026 quarter.²⁸ That is a leadership claim about volume, not about revenue quality — ETFs and index strategies supplied two-thirds of it.²⁸ A dollar into QQQ and a dollar into a Pantheon secondaries fund are not the same dollar, and conflating them is how thematic screens go wrong.

High-fee alternatives flows. AMG leads among the multi-boutique platforms, at roughly $100 billion over the trailing twelve months to June 2026.²⁵ The dossier's assertion that GQG and Artisan lead organic flow momentum does not survive the 2026 data: GQG ran roughly US$15 billion of outflows in the first half, and Artisan lost $5.7 billion from one mandate.¹⁴¹³ That claim should be retired.

Organic growth rate in a regional market. Pinnacle leads in Australia, with 18 affiliates and record first-half net inflows of A$17.2 billion.²² The lead is genuine and does not generalise — Pinnacle's advantage rests on being the default institutional partner for Australian boutiques in a market with a large, concentrated superannuation pool. It has no equivalent position in the United States. And its parent-level profit fell in the same half in which affiliate FUM rose 13%, which is the clearest available illustration that affiliate growth and shareholder returns are separate variables.²²

Cash-rate-linked scale. Federated Hermes leads on money market assets, at $676.9 billion of a record $911.6 billion total at 30 June 2026, though its money-fund market share slipped to 6.7% from 6.9% in the prior quarter.³³ Its revenue mix — 50% money market, 30% equity, 10% fixed income, 8% alternatives, private markets and multi-asset, 2% other — makes it the least pure expression of this theme among the companies here.³³ It is nonetheless moving toward the model: on 9 April 2026 it completed the purchase of 80% of FCP Fund Manager for $331 million, comprising $215.8 million in cash, $23.2 million in Class B stock and up to $92 million of contingent consideration, adding a US multifamily real estate manager with $3.5 billion of client assets and a history of investing or financing more than $14.8 billion of gross asset value.³⁴ The earnout structure is the interesting part: a fifth of the consideration is conditional on future outcomes, which is how a disciplined buyer prices key-person risk.

Put together, the field looks like this. Nobody leads everything. AMG leads on high-fee flow capture and per-share compounding. Invesco leads on volume. Pinnacle leads in one geography. Federated leads on cash-rate exposure. Artisan leads on payout. And the firm with the purest independence, GQG, is having the worst year of any of them.

↑ Back to Top

Section 7: History and Inflection Points: From the 1993 AMG Blueprint to the Private Credit Boom

The 1997 prospectus that took AMG public described a company built on an idea Wall Street found hard to underwrite: a holding company that owned economics in firms it did not control.³ The obvious objection was that if a boutique underperformed, the parent had no lever to pull. Three decades later the objection remains valid — and the model survived anyway, because the alternative levers turned out to be worse.

1993–1997: the structure. Nutt built on and departed from the United Asset Management template that preceded him, the key departure being that affiliate managers received genuine equity in their own firm rather than a salary and a corporate title.¹ Tweedy, Browne in 1997 proved the model could attract a first-rank name.¹ The November 1997 IPO raised $202 million and gave the structure a public currency for further deals.¹³

1997–2007: the golden decade. Active public equity management was still a growth business. Fees held. Boutiques with distinctive processes gathered assets. A platform that could buy into them without breaking them had a straightforward value proposition.

2008–2009: the break. The financial crisis did two things at once. It destroyed the performance of many active equity strategies in a single year, and it ushered in a decade of near-zero rates that made the cost of an active fee impossible to ignore against an index fund. The passive share of equity assets began the climb that has not stopped. Across the largest pension markets, equity allocations fell from around 60% to 43% over the two decades ending 2020, with the money going into alternatives.⁵

2010–2015: the pivot begins. AMG started buying capabilities that indexing could not commoditise — private equity and infrastructure at Pantheon, quantitative liquid alternatives at Systematica. In 2014 it took a minority stake in EIG Global Energy Partners, with EIG's management keeping majority ownership.²⁶

2016–2020: competition arrives. The GP-stakes industry professionalised. Petershill and Dyal, which had been niche vehicles in 2007 and 2011, became institutional franchises; by 2021 Dyal had merged into Blue Owl through a listed vehicle and Petershill had floated in London, and more than $20 billion was raised across GP stakes funds in the fourth quarter of 2021 alone.¹¹ For AMG this changed the acquisition market. It was no longer the only buyer offering money without control, and the price of minority stakes rose accordingly.

2021–2024: the private credit boom. Rates rose sharply, banks retreated from middle-market lending, and direct lending became the fastest-growing product in asset management. AMG's exposure came through affiliates including Comvest, and the flows followed.

2025–2026: the wealth channel opens — and the first crack. Semi-liquid evergreen vehicles brought private markets to advisers and their clients. The AMG Pantheon Fund reached approximately $4.5 billion, offering a global private equity portfolio with quarterly liquidity, within a Pantheon evergreen platform of roughly $8.6 billion.³⁵ In May 2025 AMG agreed to take a minority stake in Qualitas Energy.²⁷ Victory closed the Amundi US transaction and revived Pioneer.¹⁸ Federated bought 80% of FCP.³⁴

Then, on 19 February 2026, Blue Owl restricted withdrawals from a retail-focused fund, and asset manager shares fell across the board.³² By mid-March, reporting on the $1.8 trillion private credit market described more than $265 billion of market capitalisation erased across the listed alternative managers.³⁶ By late June, Blue Owl, Apollo and Ares had fallen further, with private-credit-heavy managers hit hardest — Apollo at 86% of fee-earning assets in private credit, Ares at 66%, Blue Owl at 53%, KKR at 48%.³⁷ For the year to that point, Apollo and Blackstone were down roughly 12%, Ares 15%, KKR close to 16%, and Blue Owl nearly 18%.³⁶

The bear case in the dossier — private market fee compression and valuation contagion — was not a hypothetical in 2026. It was the year's main event.

The regulatory story, corrected

The dossier asserts that SEC private fund adviser rules from 2023 onward forced boutiques into the arms of platforms by raising compliance costs. That is wrong, and the correction matters because it removes a load-bearing plank from the bull case.

The SEC adopted the Private Fund Adviser Rules in August 2023. On 5 June 2024 the Fifth Circuit vacated them in their entirety, holding that the Commission had exceeded its statutory authority under sections 206(4) and 211(h) of the Investment Advisers Act of 1940.³⁸ Private fund advisers have had no obligation to comply since.³⁸ The US compliance-burden tailwind that was supposed to push small managers toward platforms did not arrive.

Nor did the retirement-advice one. The Department of Labor finalised its Retirement Security Rule in April 2024; two federal district courts stayed it in July 2024; and in March 2026 the courts vacated it entirely. The Department restored the 1975 five-part test for investment-advice fiduciary status effective 20 April 2026.³⁹⁴⁰

Europe went the other way. AIFMD II took effect on 16 April 2026, and its delegation requirements now extend beyond portfolio and risk management to fund administration, marketing, loan origination and other activities, with managers required to give regulators far more detail on delegation arrangements and to demonstrate genuine substance and adequate technical and human resources.⁴¹⁴² Level 2 implementing measures were delayed until after 1 October 2027, which creates the awkward combination of binding obligations and unsettled detail.⁴²

The net institutional picture: the regulatory push toward platform membership is real in the European Union and largely absent in the United States. For a US-listed multi-boutique platform, that is a meaningfully weaker structural tailwind than the standard pitch implies — and it means the case has to rest on distribution economics and succession demand, not on compliance costs.

What management said, and when

One useful discipline with a company that reports a lot of adjusted figures is to check whether its explanations changed as the numbers did.

On the first-quarter 2026 call, AMG's management attributed the quarter's $25 billion of liquid alternatives inflows to a named group of affiliates — AQR, Capula, Garda, Systematica and Winton — and its $4 billion of private markets fundraising to Pantheon secondaries plus infrastructure raises at Arra, EIG and Qualitas Energy.¹⁷ That is a specific answer. It names the sources, distinguishes liquid from illiquid, and can be checked against subsequent quarters.

By the second quarter, with alternatives inflows at $29 billion and roughly $100 billion over twelve months, the framing shifted to the earnings mix: alternatives above 60% of earnings, up from about 50% eighteen months earlier.²⁵ Also specific, and consistent with the 58% of EBITDA figure disclosed a quarter earlier.²⁴ Chief Executive Jay C. Horgen's own summary of the quarter led with the growth rates in adjusted EBITDA and economic earnings per share rather than with assets under management — 44% and 54% respectively — which is a reasonable emphasis for a company whose asset base and earnings base have diverged, and also a reminder that management chooses the frame.⁴

The consistent thread across both calls is that AMG has been willing to name where flows came from, which is more than many peers do. The persistent gap is affiliate-level profitability, which the company does not disclose on confidentiality grounds. That gap is not evasion — the confidentiality obligation is real, and it is part of what makes founders willing to sell — but it means an outside investor is trusting an aggregate they cannot decompose.

↑ Back to Top

Section 8: Public Market Pricing, Valuation Dispersion, and the Expectations Variant Wedge

Here is the accounting problem that has defined how this company is valued.

When a holding company owns a minority position in an affiliate — broadly between 20% and 50% — it does not consolidate that affiliate's revenue. Only its share of the affiliate's net income appears, on one line, as equity in earnings of unconsolidated affiliates. So as AMG shifted capital into minority private-markets stakes, its reported revenue line stopped reflecting the size of the business underneath it.

The gap is visible in a single quarter. For the three months to June 2026, AMG reported consolidated revenue of $640.7 million, GAAP net income attributable to controlling interests of $185.9 million and diluted GAAP earnings of $6.95 a share — against economic net income of $221.4 million and economic earnings per share of $8.29.⁴ Economic net income adds back non-cash items, principally intangible amortisation and certain deferred tax effects, on the argument that they do not consume cash available for reinvestment or repurchase. That argument is reasonable. It is also a management-defined measure, and a sceptic is entitled to note that amortisation of an intangible representing a purchased revenue stream is not obviously a non-economic charge when the revenue stream can decay.

For most of the past decade, the market split the difference by simply refusing to pay up. That is no longer true.

Exhibit 3 — Valuation dispersion, close of 31 July 2026 Definition: closing price, market capitalisation, trailing GAAP price/earnings, and drawdown from the trailing 52-week high. Geography: US listings. Evidence status: market data; forward multiples where shown are third-party estimates on differing earnings definitions and are not comparable with trailing GAAP figures.

Company	Price ($)	Mkt cap ($B)	Trailing P/E	Drawdown from 52-wk high
AMG	366.73	9.69	12.7	−4% (high $382.75)
Victory Capital	99.03	6.19	22.3	−3% (high $102.05)
Federated Hermes	59.96	4.27	11.2	−2% (high $61.01)
Franklin Resources	33.86	17.20	23.0	−3% (high $34.92)
T. Rowe Price	111.75	23.84	11.2	−8% (high $122.00)
Artisan Partners	39.24	3.18	9.4	−19% (high $48.46)
Blackstone	127.74	158.90	28.6	−33% (high $190.09)
KKR	101.43	91.06	32.4	−33% (high $152.10)
StepStone	43.86	5.34	n/a	−44% (high $77.80)
Hamilton Lane	88.91	4.96	15.0	−45% (high $161.13)

Sources: consolidated closing market data, 31 July 2026.⁴³⁴⁴ AMG's 52-week low was $200.00 and its 200-day average $299.63.⁴³

Read that column of drawdowns aloud and the dossier's central premise inverts. The proposition was that multi-boutique platforms trade at traditional-manager multiples of nine to eleven times while pure-play alternative managers command eighteen to twenty-five times, leaving a gap for someone to arbitrage. As of 31 July 2026, AMG sat within 4% of its 52-week high after roughly doubling off a $200 low, while Hamilton Lane sat 45% below its high and StepStone 44% below.⁴³⁴⁴ Blackstone and KKR were each a third below their highs.⁴³⁴⁴ The gap did not close because multi-boutiques re-rated up to alternatives managers. It closed from both directions, and the alternatives side did most of the moving.

On AMG's own economic measure, the arithmetic runs roughly as follows. Economic EPS was $26.05 for 2025, $8.23 in the first quarter of 2026 and $8.29 in the second, with third-quarter guidance of $8.43 to $8.71.²⁴⁴²⁵ That implies a trailing-twelve-month figure near $32 and a 2026 full-year figure in the mid-thirties. At $366.73 that is roughly eleven times trailing and near ten and a half times the current-year run rate — a real multiple, and no longer the eight-to-nine times the thesis was built on. Third-party data showed a trailing GAAP P/E of 12.7 and a forward multiple of 9.2, with an analyst consensus target of $429 and a "Strong Buy" designation; those are consensus data points, not facts about value.⁴³

Comparing AMG to Blackstone and KKR on headline multiples is treacherous. Their trailing GAAP P/Es of 28.6 and 32.4 sit against third-party forward estimates of roughly 17.3 and 13.8 — a divergence driven by the gap between GAAP net income and the distributable-earnings measures those firms emphasise.⁴⁴³¹ Any comparison across these companies must pick one earnings definition and hold it, and no public source reviewed here does so cleanly across all of them. Morningstar's coverage in mid-2026 noted that most traditional and alternative managers it tracked were trading below its fair value estimates, with discounts reaching about 15%.⁴⁵

What the price now requires

So what does an investor need to believe at $366.73?

At roughly ten and a half times current-year economic earnings, the price does not require heroic growth. It requires the alternatives franchise to keep growing at something like recent rates, the long-only drag to remain a drag rather than becoming a collapse, and the buyback to continue at scale. Management's stated intention was roughly $600 million of repurchases in 2026 against a $9.69 billion market capitalisation — a little over 6% of shares at current prices.²⁵⁴³

There is a mechanism here that deserves more attention than it gets, because it works against the bulls. The buyback's accretive power is a function of the multiple. Retiring stock at eight times earnings adds far more per share than retiring it at fourteen. AMG's five-year share count reduction of 37% was achieved substantially while the stock was cheap.²⁴ Now that it is not, the same dollars buy fewer shares and add less. A re-rating is good for the holder who already owns it and worse for the compounding engine going forward. Bulls who want both the re-rating and the historical accretion rate are asking for two things that partly cancel.

The sceptic's list

A serious long/short investor would press on five points.

First, quality of the alternatives earnings. Performance fees were $49 million in the first quarter of 2026, up $29 million year on year, against quarterly adjusted EBITDA of roughly $316 million.²⁴⁴ Some meaningful part of the recent growth is a good year for performance fees, which are cyclical and do not deserve a structural multiple.

Second, the non-GAAP bridge. Economic EPS exceeded GAAP EPS by $1.34 in the second quarter of 2026 — about 16%.⁴ That is not egregious, but it is the direction of travel that matters if the adjustments grow.

Third, opacity. AMG does not disclose affiliate-level margins or profitability, citing confidentiality with its partners. An investor therefore cannot see whether the alternatives strength is broad or concentrated in two firms. Concentration risk of that kind is unmeasurable from outside, and the honest response is to widen the required margin of safety rather than to assume it away.

Fourth, the long-only book. Roughly $474 billion of differentiated long-only assets at March 2026 lost something like $16 billion net in the June quarter.²⁴⁴ Each departing dollar reduces the operating pool at affiliates whose fixed costs do not fall in step. That is the renegotiation-and-impairment risk described earlier, and it is a slow fuse rather than a headline event.

Fifth, the sector's own reflexivity. AMG's stock roughly doubled off its 52-week low while its private-markets comparables fell by more than 40%.⁴³⁴⁴ Part of that is genuine operating outperformance. Part is that AMG's private credit exposure is smaller and more diversified than a pure-play lender's. And part may simply be that a stock trading at eight times had further to run than one trading at twenty-five. Distinguishing the three from outside is not possible with the evidence available, and an investor should say so.

↑ Back to Top

Section 9: The Falsification Dashboard: Scenarios, Monitoring Indicators, and Crux KPIs

A thesis worth holding is one you know how to abandon. Here are the three worlds, each with a distinct causal path rather than a different number.

The bull world runs on the wealth channel. Semi-liquid evergreen vehicles become a standard 5–10% allocation across RIA and private bank model portfolios. Demand shifts from a fundraising cycle to a subscription flow, which changes the character of the revenue from lumpy to recurring. Alternatives inflows sustain at $80–100 billion a year across AMG's affiliates, alternatives push past two-thirds of earnings, performance-fee-eligible assets grow with the base, and the buyback continues. Competition for stakes stays rational because GP-stakes funds are absorbed digesting 2021-vintage deals. In this world, the constraint that binds is distribution capacity, and the winners are platforms with existing wealth shelf space.

The base world is the grind. Alternatives inflows moderate to $40–60 billion a year, enough to offset long-only redemptions but not to accelerate. Performance fees normalise downward from a strong 2025–26. Economic earnings grow at high single digits, roughly two-thirds of it from repurchase and one-third from the business. The multiple stays where it is. The constraint is the long-only decay rate, and the outcome depends almost entirely on whether that decay stays linear.

The bear world is a private markets credit cycle. Defaults rise in middle-market direct lending. Marks fall. Performance fees stop crystallising entirely and some accrued carry reverses. Evergreen vehicles hit redemption limits, as Blue Owl's retail fund did in February 2026, and wealth-channel adoption stalls for two to three years while advisers explain the gates to clients.³² Fundraising cycles stretch from eighteen months to thirty. Simultaneously, long-only redemptions accelerate in a risk-off market, squeezing affiliate operating pools and forcing revenue-share renegotiations. The constraint is liquidity, and it binds at both ends of the barbell at once.

Note that the bear world is not speculative. Its first act happened in the first half of 2026, and the alternatives complex lost more than $265 billion of market value.³⁶ What has not yet happened is the second act, in which the credit losses actually arrive.

The crux KPIs

Five observables discriminate between these worlds. Each leads rather than lags, each sits on a binding constraint, and each has a named source and a threshold.

Exhibit 4 — Crux KPIs Evidence status: latest readings are company-disclosed; thresholds are analytical judgments stated for falsification, not forecasts.

KPI	Latest reading	Why it leads	Confirm / break
Alternative-strategy net client cash flows	+$29B (Q2 26); ~$100B LTM425	Allocator commitments precede fee revenue by 2–6 quarters	Break the theme below +$40B LTM; break decisively on two consecutive quarters of net alternatives outflows
Performance-fee-eligible AUM and its correlation mix	$240B eligible; 75% low or negative correlation to public markets (Q1 26)24	Sets the option value of future carry before any is earned	Break below ~$200B eligible, or low-correlation share under 60%
Evergreen vehicle subscriptions and gate usage	AMG Pantheon Fund ~$4.5B with quarterly liquidity; Pantheon evergreen ~$8.6B35	Wealth adoption is the whole bull path; gating is the fastest possible reversal	Break on any AMG-affiliated evergreen vehicle limiting redemptions
Annualised share-count reduction	43M → 27M over five years to Mar 26 (−37%); $375M repurchased H1 26244	Determines per-share outcome independently of the business	Security-level break below 3% a year, or capital diverted to stakes above ~12x
Alternatives affiliate benchmark outperformance	84% of private markets AUM ahead of benchmark, latest vintage; 92% of liquid alts ahead over 3, 5 and 10 years (Q1 26)24	Consultant reviews follow performance by roughly a year; flows follow reviews	Break below 60% of alternatives AUM outperforming

Spoken plainly: the first three tell you whether the theme is alive, the fourth tells you whether the shareholder gets paid for it, and the fifth tells you whether the first three will still be true in two years. The fourth is the one that separates a correct industry call from a good stock outcome — a platform can capture the flows and still deliver a mediocre return if it stops retiring stock or overpays for the next stake.

All five come from quarterly company disclosure — earnings releases, 10-Qs and investor supplements — with the exception of the outperformance figures, which are company-compiled against affiliate-specific benchmarks and therefore self-reported. Independent verification of manager performance requires consultant databases that are not publicly available, which is a real limitation and should be stated rather than papered over.

Distinguish, finally, between two kinds of failure. Theme kill: alternatives inflows across the industry turn negative for a year, or evergreen gating becomes systemic — at which point the migration of profit pools into private markets has stalled and the whole construct is wrong. Security kill: the theme holds but AMG stops converting it into per-share value, through a diluted acquisition, an abandoned buyback, or an affiliate impairment large enough to reset the earnings base.

Where the hidden bets are

Anyone building exposure here should know what they are duplicating.

The obvious shared factor is equity beta through mark-to-market on assets. The less obvious one is credit. AMG's alternatives earnings, StepStone's and Hamilton Lane's fee bases, and the entire listed private-markets complex all lean on the same private credit and private equity valuation regime. Owning a multi-boutique platform alongside a private-markets specialist is one bet expressed twice, and 2026 demonstrated that the two move together when stressed.³⁶³⁷

There is a rate bet in both directions. Higher short rates lift Federated Hermes' money market franchise, which at $676.9 billion is over 70% of its assets and half its revenue.³³ Higher long rates raise discount rates on private assets and lengthen exit timelines, which hurts everyone else.

And there is single-name concentration hiding inside a diversified-sounding structure. AMG's first-quarter liquid alternatives inflows of $25 billion came from a named handful of firms.¹⁷ Affiliate-level economics are undisclosed. It is entirely possible that a large share of the platform's alternatives earnings comes from two or three affiliates, and no outside investor can rule that out.

The offsetting pathways are worth naming analytically, without turning them into trade ideas. Invesco's index franchise gains from precisely the fee compression that hurts long-only affiliates.²⁸ Federated's cash franchise gains from the rate environment that pressures private asset valuations.³³ These are structurally opposed exposures within the same sector, which is the reason a list of asset-management tickers is not a diversified position.

↑ Back to Top

Section 10: Future Game Changers, Wealth Democratization, and Value Migration

Three developments could change where the profit pool sits over the next five years. Each has a mechanism, a hurdle, and a milestone you can actually watch.

The wealth channel, and the liquidity problem it created

The mechanism is straightforward. A closed-end institutional fund requires a subscription document, a multi-million-dollar minimum, capital calls over years, and a ten-year lock. That is unsellable to an adviser managing a hundred client relationships. An evergreen vehicle takes continuous subscriptions and offers periodic redemptions, usually quarterly, subject to a cap. It converts a private markets fund into something an adviser can put in a model portfolio.

The scale is already real rather than promised. The AMG Pantheon Fund holds approximately $4.5 billion, offering a global private equity portfolio sourced across Pantheon's platform with quarterly liquidity, inside a Pantheon evergreen range of roughly $8.6 billion globally, developed and distributed with AMG.³⁵ StepStone's private wealth platform approached $18 billion by the close of its 2026 fiscal year.²⁹ These are meaningful businesses.

The hurdle is the thing that broke in February 2026. An evergreen fund promises periodic liquidity on an illiquid portfolio. That promise holds until enough investors test it at once. Blue Owl's decision to restrict withdrawals from a retail-focused fund, after redemptions in a related vehicle ran roughly 20% above the prior year, sent asset manager shares down across the sector.³² The structural point is that this is a shared-fate mechanism. An adviser who watched one sponsor gate a fund will ask harder questions about every sponsor's fund, regardless of the underlying portfolio.

Beneficiaries if adoption resumes: platforms with existing wealth distribution and institutional-quality private markets product — AMG through Pantheon, EIG and Comvest; StepStone and Hamilton Lane directly. Losers: sub-scale private managers without a distribution partner, who will find the wealth channel closed to them, and traditional long-only houses whose model-portfolio shelf space the evergreen products take.

Observable milestone: quarterly net subscriptions to the named evergreen vehicles, and any use of redemption limits. That is the single highest-information data point in this theme.

Tokenisation

The mechanism is administrative rather than financial. Representing fund interests as digital tokens can cut transfer-agency cost, enable secondary transfer without a bilateral negotiation, and lower minimum subscriptions. PwC's research projected tokenised products growing from about $40 billion to more than $317 billion by 2028, a 51% compound rate, with managers planning to offer tokenisation most often in private equity, listed equity and hedge funds.⁸ That is a vendor-adjacent forecast and should be read as one — the observed base is small.

The hurdle is that the binding constraint on private-market liquidity is the underlying assets, not the register. A token representing an interest in a portfolio of unlisted companies is still an interest in a portfolio of unlisted companies. Tokenisation can make transfer cheaper; it cannot make a buyout stake liquid. Anyone selling it as a liquidity solution is selling the wrong thing, and the gap between announced pilots and scaled commercial adoption remains wide.

Where value could move: toward transfer agents, administrators and platform providers that own the token infrastructure, and slightly away from managers, if easier secondary transfer erodes the lock-up premium that justifies part of the fee.

Re-rating by reclassification

The third possibility is the one the sell side likes and the evidence supports least. If index providers or rating agencies reclassified multi-boutique holding companies as alternative managers, passive flows would follow. The mechanism is real. The hurdle is that AMG's market capitalisation is $9.69 billion against Blackstone's $158.90 billion, and index reclassification at that size moves a modest amount of money.⁴³⁴⁴ More to the point, the 2026 evidence suggests the market already repriced AMG on fundamentals — the stock roughly doubled off its low while the alternatives complex fell.⁴³⁴⁴ A classification change would be confirming a move that has largely happened.

Where a rotation would come from

If the theme continues to work, the interesting question becomes which layer captures it, and the answer changes with price.

Through 2024 and 2025 the multi-boutique layer was the cheap way to own alternatives growth, and that argument has largely been paid out — AMG within 4% of its high, the private-markets specialists 44% and 45% below theirs.⁴³⁴⁴ The mechanical consequence is that the relative-value case now points the other way, toward the layer that de-rated. Whether that is an opportunity or a warning depends entirely on whether the private credit stress that caused the de-rating turns into realised losses. If it does, the specialists are cheap for a reason and the multi-boutique platforms will follow them down with a lag, because their alternatives earnings sit on the same asset marks. If it does not, the de-rated layer has more to recover.

That is the honest framing of a rotation: a conditional statement about which layer carries the risk, not a recommendation about which ticker to hold. And it carries a warning about false diversification. A portfolio containing a multi-boutique platform, a private-markets specialist and a listed alternatives manager holds three expressions of one bet on private asset valuations. The 2026 drawdowns demonstrated the correlation directly.³⁶³⁷

The downside fault lines

Two risks dominate, and they are correlated.

Key-person loss at a flagship affiliate. AMG cannot prevent it, by construction. The propagation path is consultant review, then platform freeze, then redemption, over roughly a year. Because affiliate-level earnings are undisclosed, the market cannot price the risk until the announcement, which means the repricing is discontinuous.

Private markets valuation contagion. The first half of 2026 provided the template: a liquidity event at one sponsor, sector-wide repricing, more than $265 billion of listed market value erased, and private-credit-heavy managers hit hardest.³⁶³⁷ What has not yet happened is a genuine default cycle in middle-market direct lending. If it comes, it hits performance fees, fundraising, evergreen redemptions and affiliate stake valuations simultaneously — and the platforms with the most alternatives earnings, which is now the bull case, would have the most exposed profit and loss.

There is also a capital-cycle risk that gets little attention. GP-stakes funds, listed alternatives firms and multi-boutique platforms are all bidding for the same finite population of good mid-sized managers. When several well-capitalised buyers chase one scarce asset class, entry multiples rise and forward returns on deployed capital fall. Atlas Holdings taking simultaneous GP-stakes backing from Blackstone and Blue Owl in March 2026 is a sign of a competitive, not a distressed, market for these stakes.¹² For a platform whose returns depend on buying stakes at attractive multiples, a competitive market is a headwind, and it argues for buying back stock over buying more affiliates — which is, notably, what AMG has been doing.

What the upstream belief looks like now

The proposition we started with had two halves. Test them separately, because they have not aged the same way.

The bifurcation half is holding, and arguably strengthening — but not in the shape the theme predicted. In the June 2026 quarter, the barbell's two ends both won: Invesco's index and ETF franchise took $30.1 billion of a record $45.1 billion, while AMG's alternatives affiliates took $29 billion.²⁸⁴ The squeezed middle is real. Undifferentiated active long-only management is losing assets at AMG, at T. Rowe Price, and at Artisan.⁴¹⁶¹³ What the theme got wrong was the relative volume. The passive end is capturing far more dollars; the specialist end is capturing far more revenue per dollar. Both statements are true and they lead to different investment conclusions.

The talent half is fraying. The claim that centralised managers structurally cannot retain elite talent while independent boutiques can is too strong. Franklin's Western Asset drag supports it.⁹ But Artisan — decentralised, autonomous, well-governed — lost a strategy to one client decision, and GQG — fully independent, founder-owned, answerable to no parent — ran roughly US$15 billion of outflows in six months.¹³¹⁴¹⁵ Structure protects against self-inflicted damage. It does not protect against the market deciding your strategy no longer works.

And the corollary that made this a trade — that public markets systematically misprice the model — has largely been arbitraged away. AMG traded at $366.73 on 31 July 2026, within 4% of a 52-week high, roughly 22% above its 200-day average, having roughly doubled from $200.⁴³ The pure-play private-markets names it was supposed to be cheap against were 33% to 45% below their highs.⁴³⁴⁴ The gap that the thesis was built on closed in about twelve months, and most of the closing came from the other side falling.

That leaves an investor with a cleaner, harder question than the one the theme started with. The structural case for the architecture is sound: buying economics while leaving control with the people who generate them is a genuinely better way to own investment franchises, and the flow data supports it. The compounding case has become more demanding, because a buyback is a weaker engine at eleven times earnings than at eight. And the risk case has moved from theoretical to observed, because the private markets stress the bear scenario described arrived in February 2026 and has not resolved.

Correct about the industry, and still able to lose money on the security. That was the distinction we set out with, and it is the one this year has been teaching.

↑ Back to Top

Glossary

Affiliate. An independent investment management firm in which a multi-boutique platform holds a minority or majority economic interest while the firm keeps its own brand, investment process and hiring. The affiliate, not the platform, employs the people who generate returns.

Structured partnership interest / revenue-sharing agreement. AMG's core contract form, under which an affiliate allocates a specified percentage of its revenue to AMG and to affiliate management, with the remainder funding operating expenses and further distributions to affiliate management.² For most affiliates AMG shares in revenue without regard to expenses, which insulates its economics from affiliate cost inflation and concentrates the pain of declining revenue on the partners' pool.

Operating allocation. The portion of an affiliate's revenue retained to pay salaries, bonuses, research, technology and premises. Its adequacy relative to competitive pay is the single best predictor of whether an affiliate's key people stay.

Owners' allocation. The remaining share of revenue divided between the platform and the affiliate's partners. This is what a multi-boutique platform is actually buying.

Economic net income / economic earnings per share. A company-defined, non-GAAP measure adding back non-cash charges — principally intangible amortisation and certain deferred tax effects — to GAAP earnings. It matters because equity-method accounting understates the scale of minority-owned affiliates, and it is contestable because amortisation of a purchased revenue stream is not obviously non-economic.

Equity-method accounting. The treatment applied when an owner holds roughly 20–50% of a business: the affiliate's revenue never appears on the owner's income statement, only the owner's share of its net income. This is why revenue-per-dollar-of-AUM calculations are misleading for multi-boutique platforms.

GP staking. Buying a permanent minority interest, typically 10–30%, in an alternative manager's general partnership, entitling the buyer to a pro-rata share of management fees, balance-sheet returns and carried interest.¹⁰ The main competing bidder for the assets multi-boutique platforms want.

Evergreen fund. A semi-liquid vehicle taking continuous subscriptions and offering periodic redemptions, usually quarterly and usually capped. It is how private markets reach wealth channels, and the redemption cap is both its enabling feature and its principal risk.

Gate. A contractual limit on redemptions in a given period. When one sponsor uses one, the entire product category is repriced, as happened in February 2026.³²

Performance-fee-eligible AUM. Assets on which a manager could earn incentive fees if hurdles are met. It is a leading indicator of future carry, distinct from carry actually crystallised, and its correlation profile to public markets determines how reliable that future carry is.

Net client cash flows. Subscriptions minus redemptions, excluding market movement. The cleanest available measure of whether clients are choosing a manager, and the reason a rising AUM figure can conceal a deteriorating franchise.

Gatekeeper. An institutional consultant, wirehouse platform, private bank or RIA investment committee that decides which managers may be considered. Gatekeepers screen on operational infrastructure and firm stability as much as returns, which is what makes platform membership valuable to a small manager.

Differentiated long-only. A manager's term for active public-market strategies it argues are distinct enough to resist index competition. At AMG it was $474 billion of assets and a minority of profit at March 2026 — the part of the business the pivot was designed to outgrow.²⁴

References

1. Affiliated Managers Group, Inc. — Company Profile, Information, Business Description, History — Reference for Business ↩↩↩↩↩↩↩↩

2. Affiliated Managers Group, Inc. — 2020 Annual Report ↩↩↩↩↩↩↩↩↩↩

3. Affiliated Managers Group Inc — Form 424B4 (initial public offering prospectus), 1997 — U.S. Securities and Exchange Commission ↩↩↩

4. AMG Reports Financial and Operating Results for the Second Quarter and First Half of 2026 — GlobeNewswire, 30 July 2026 ↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩

5. Global pension funds weather the storm of 2020 — Willis Towers Watson, February 2021 ↩↩↩

6. Global pension assets climb to record $58.5 trillion — WTW, February 2025 ↩

7. Global Pension Assets Study 2025 — Thinking Ahead Institute ↩

8. PwC 2024 Asset & Wealth Management Report — PwC ↩↩

9. Franklin Resources, Inc. Announces Preliminary Month-End Assets Under Management — Business Wire, 6 July 2026 ↩↩↩↩↩↩↩

10. What is GP stakes investing? — PitchBook ↩↩

11. With Dyal and Petershill Going Public, Investcorp's Fundraise Shows Growing Appetite for Smaller Alternative Firms — Institutional Investor ↩↩↩↩↩↩

12. Atlas Holdings lands dual GP stakes backing from Blackstone, Blue Owl — Alternatives Watch, 11 March 2026 ↩↩↩

13. Artisan Partners Asset Management Inc. Reports June 2026 Assets Under Management — GlobeNewswire, 10 July 2026 ↩↩↩↩↩↩↩

14. Outflows and investment performance weigh on GQG Partners — Money Management ↩↩↩↩

15. GQG Partners (ASX:GQG) Posts Sixth Consecutive Month of Net Outflows in 2026 — Kalkine Media ↩↩↩

16. T. Rowe Price Group Reports Month-End Assets Under Management for June 2026 — PR Newswire, 13 July 2026 ↩↩↩↩↩

17. Affiliated Managers Group Q1 2026 Earnings Call: Complete Transcript — Benzinga, May 2026 ↩↩↩↩↩

18. About Victory Capital — Victory Capital ↩↩↩

19. Victory Capital Reports June 2026 Total Client Assets — Victory Capital Holdings, July 2026 ↩↩

20. Amundi Q2 2026 slides: record profit, AuM hit €2.6tn — Investing.com, July 2026 ↩

21. Artisan Partners Asset Management Inc. — Form 8-K, second quarter 2026 press release — U.S. Securities and Exchange Commission, July 2026 ↩↩↩

22. Earnings call transcript: Pinnacle Investment sees profit drop in H1 2026 — Investing.com ↩↩↩↩↩↩

23. Pacific Current Group's FUM Hits A$26.4b as Continuing Boutiques Grow 1.8% After Aether Exit — Kalkine Media, July 2026 ↩

24. AMG Q1 2026 slides: record AUM, 58% earnings growth on alternatives — Investing.com, May 2026 ↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩

25. Earnings call transcript: AMG posts record Q2 2026 results as AUM hits $942 billion — Investing.com, July 2026 ↩↩↩↩↩↩↩↩↩↩↩↩↩

26. AMG Buys Stake in EIG Global Energy as Sonneborn Joins — Bloomberg, 28 March 2014 ↩↩↩↩

27. AMG and Qualitas Energy Announce Partnership — Affiliated Managers Group, May 2025 ↩↩↩↩↩

28. Invesco Q2 2026 slides: record $45B inflows drive margin expansion — Investing.com, July 2026 ↩↩↩↩↩↩↩↩↩

29. StepStone caps record fiscal year as private wealth platform nears $18bn — Alternatives Watch, 21 May 2026 ↩↩↩

30. Hamilton Lane Incorporated Reports Third Quarter Fiscal 2026 Results — Hamilton Lane ↩↩

31. BX vs KKR comparison tool — Tickeron ↩↩

32. Asset Manager Stocks Slide After Blue Owl Restricts Fund Withdrawals — Bloomberg, 19 February 2026 ↩↩↩↩↩↩

33. Federated Hermes, Inc. reports record assets under management with second quarter 2026 earnings — PR Newswire, 30 July 2026 ↩↩↩↩↩

34. Federated Hermes enters U.S. real estate market by acquiring majority stake in FCP Fund Manager — Pensions & Investments, 2026 ↩↩

35. Pantheon expands US private wealth offerings with evergreen infrastructure fund — Pantheon ↩↩↩

36. The $265 billion private credit meltdown: How Wall Street's hottest investment craze turned into a panic — Fortune, 14 March 2026 ↩↩↩↩↩↩

37. Private Credit Turmoil Hits Blue Owl, Apollo Global and Ares As Their Stocks Sink — Benzinga, June 2026 ↩↩↩↩

38. Fifth Circuit Vacates SEC Private Fund Adviser Rules in Full — Morgan Lewis, June 2024 ↩↩

39. US Department of Labor restores long-standing investment advice rule after pair of court decisions vacate 2024 retirement security rule — U.S. Department of Labor, 18 March 2026 ↩

40. Retirement Security Rule: Definition of an Investment Advice Fiduciary — Notice of Court Vacatur — Federal Register, 20 March 2026 ↩

41. 16 April 2026: AIFMD II Finally Takes Effect — Chambers and Partners ↩

42. AIFMD II Roundup: Key Reforms, EU Implementation and the UK Alternative — Skadden, Arps, Slate, Meagher & Flom, April 2026 ↩↩

43. Affiliated Managers Group (AMG) stock overview — Stock Analysis, 31 July 2026 ↩↩↩↩↩↩↩↩↩↩↩↩

44. Stock comparison: AMG, VCTR, APAM, FHI, TROW, BEN, IVZ, STEP, HLNE, BX, KKR — Stock Analysis, 31 July 2026 ↩↩↩↩↩↩↩↩↩

45. Why Alts Manager Stocks Are Getting Hit Hard — Morningstar, 2026 ↩