The Cheap Genius Problem

A fund manager generated US$13 billion of investment performance for its clients in the first four months of 2026. The clients responded by pulling US$9.9 billion out of the firm.

That sentence is worth enjoying for a minute before you decide what to do with it.

The fund manager is GQG Partners (ASX:GQG). The numbers above are not from a deck or a sell-side note. They come from GQG's own monthly funds-under-management update for April 2026, signed off by the company. The stock trades around A$1.57 to A$1.59. The trailing P/E is roughly 6.6×. The forward distribution yield is somewhere between twelve and fourteen percent, depending on which exchange rate you torture into giving you an answer.

This is the Cheap Genius Problem. The market has decided that the next twelve months matter more than the next ten years, the popularity contest matters more than the scoreboard, and the *direction* of flows matters more than the *magnitude* of profits. That is a very normal mistake. The trick is figuring out whether it produces a forty-cent dollar, or just a fifty-cent dollar still on its way to thirty cents.

A Six-and-a-Half-Times Earnings Multiple Walks Into a Bar

The line items, before any commentary.

- Funds under management: US$166.9bn at end-April 2026.

- Revenue (FY25): US$808m.

- Net profit (FY25): US$463m.

- Operating margin: 77%. Yes, you read that right.

- Net cash: US$133m. Debt: zero.

- Dividend policy: ~90% of distributable earnings paid out, quarterly.

- Insider ownership: ~70% sitting in QVFT (the Jain-associated holding vehicle), plus another four-odd percent in employee hands.

- Liquidity: ~A$7m a day in turnover. Spread roughly 0.6%. Included in the S&P/ASX 200 since September 2025.

The opinion comes when you decide what those numbers are worth.

At today's price, the market is paying roughly 1.8% of FUM for GQG. Mature active managers historically transact in a band several multiples of that, when allocators are in love with them. So either the market thinks the FUM number is about to halve, or it thinks the *quality* of the FUM has rotted. Both deserve to be argued. Neither has been proved.

Invert First. Always.

When a price looks this generous, the responsible thing is not to celebrate. The responsible thing is to start with the question: *what does the buyer at A$1.58 have to be wrong about for this to end badly?*

Five risks worth naming.

One. Rajiv Jain steps off the bus. Jain *is* GQG. He founded the firm. He runs the investment process. He owns about 70% of the equity through QVFT. The company says it has succession planning in place. What it has not done is publicly bolt a named heir onto the masthead. Key-person risk in a one-CIO shop is real, and it should never be argued away with words like "process" or "team." If Jain leaves involuntarily, this gets re-rated lower, faster than anyone is prepared for.

Two. The strategy stays out of fashion. GQG's Emerging Markets sleeve lagged the MSCI EM Net index by roughly twenty-one percentage points in the year to September 2025. That is not a polite stumble. That is a "you bet against AI and AI kept going" outcome. If the next twelve months look like the last twelve months, the outflows continue and FY26 earnings drift down rather than stabilise. The base case in my head assumes some recovery. I have no proof it will arrive on time.

Three. The Adani overhang is still in the room. When the US Department of Justice indicted Adani-related parties in November 2024, GQG fell roughly nineteen percent in a single session because its positions in Adani names were both large and visible. Those positions were publicly defended at the time. They have not, to my reading, been catastrophically wrong since. But the option that they could become catastrophically wrong has not been retired.

Four. The AUD strengthens. GQG earns in US dollars and pays Australian holders in AUD-converted dividends. A model assumption of AUD/USD at 0.65 is just that, an assumption. A move to 0.75 takes about thirty cents off the share's modelled fair value via translation alone. This is the most under-appreciated structural drag, and it never makes the headlines.

Five. A seventy-percent shareholder is a governance flag, not a feature. Yes, Jain's alignment is unusual and welcome. Yes, his on-market buying around A$1.70 to A$1.77 in September 2025 was a real signal. Concentration of ownership has two sides though. The day a controlling holder decides the company exists to serve a goal other than minority shareholder value, the minority shareholders learn it the slow way.

Charlie Munger had a useful sentence for this kind of inventory: *"All I want to know is where I'm going to die, so I'll never go there."* I don't think any of the five items above is fatal individually. I do think any *two* of them firing together would put the stock through the bear-case floor, and then through the floor below it.

Show Me the Trust Deed, I'll Show You the Outcome

So why even look at this thing?

Because the incentives are arranged in a way you almost never see in a public asset manager.

The same controlling shareholder who creates the governance flag also takes 70% of every dividend the company declares. The dividend policy is to pay out ninety cents of every dollar the business generates. Translate that into cash flow. Jain personally takes home roughly sixty-three cents of every after-tax dollar this company produces. He is not building a kingdom. He is not buying jets through a related-party transaction. He is, as far as the public record shows, running a profitable distribution engine and posting himself the largest cheque.

Layer on what he was doing personally in September 2025. Eleven separate on-market purchases of his own company's stock, in a price band that is roughly where it sits today. The man is eating his own cooking, in public, with witnesses.

Cooking, in this case, that comes out of a kitchen with a 77% operating margin.

This is the part of the story the bears don't talk about, because it's harder to make 77% sound bad than to make "outflows" sound bad. Asset managers that scale earn operating margins like this only when their cost base is genuinely fixed and the strategies are genuinely portable. GQG's costs are mostly people, and management has publicly said the hiring is largely done. Incremental FUM falls almost directly to the bottom line. Conversely, the firm can absorb a meaningful amount of FUM erosion before margins are seriously threatened.

You can argue about how durable the FUM base is. You cannot argue about what the existing FUM base is producing while it is here.

What's It Worth, Honestly

This is the part where most reports start lying to you with precision. I will try not to.

A ten-year DCF of distributable earnings, assuming flow drag through FY27, modest recovery thereafter, fee yield trickling down a basis point or two, and an exit multiple of ten times. That gets you to a base case fair value around A$2.45 per share. That is a thirty-five percent discount to the current price. Cross-check with a six-and-a-half-times trailing P/E on a 77%-margin business throwing off ninety percent of earnings as cash, and the same number broadly emerges from a different angle.

A bull case in which flows turn positive in the back half of 2026, the EM strategy mean-reverts, and the multiple re-rates back to twelve times. That gets you to A$3.75. A double in five years, *before* dividends. I'm not betting on it. I'm leaving room for it.

A bear case. Sustained outflows of US$25bn over the next eighteen months, an EM strategy formally demoted by allocators, key-person transition, a multiple cap at eight times, WACC raised to twelve percent. It clears at A$1.46. Roughly eight percent below the current quote.

Look at that bear case for a long minute. It does not include catastrophic Jain departure. It does not include an Adani-style headline. It is the bear case for a slow, normal, disappointing two years. And it is only eight percent below today's price.

That is the wrinkle I keep returning to. The asymmetry isn't extreme. The downside in a slow grind is small. The downside in a *real* accident, where Jain is gone, or a second Adani lands, or the AUD rips to 75, is materially worse than that A$1.46 line. The model's bear case is honest about what it's modelling. It just isn't modelling everything that could go wrong.

So when somebody tells you the discount-to-fair-value is thirty-five percent, ask which bad scenarios are inside the bear case, and which ones are sitting outside it waiting.

A Few Things That Don't Belong in the Appendix

- The dividend is completely unfranked. GQG is Delaware-incorporated. For an Australian SMSF investor used to grossed-up yields from Pinnacle or Perpetual, you do not get that arithmetic here. The headline yield is closer to the true yield than usual, which is also less attractive than usual.

- The instrument is technically a CHESS Depositary Interest (CDI) over US common stock, 1:1. This is fine, but it has tax consequences (US withholding tax, treaty reduction with a W-8BEN, no franking ever) that retail investors should not learn about in the middle of a dividend payment.

- The SEC paid the firm a US$500,000 visit in September 2024 over restrictive non-disclosure language. It is a real governance dent. Whether it materially drove the institutional outflows is an open question. I have not seen evidence either way, and I am not going to invent the link.

These are the kind of items that usually live in the appendix of a banker's pitchbook, in a font small enough to require a magnifying glass. They should not be in the appendix. They belong in the second paragraph.

A Toe, Not a Leg

So what does someone with a Munger temperament actually do here?

You do not load the boat. You do not declare it a generational opportunity. You do not write a thread on social media about the "asymmetric setup." You also do not sit on your hands forever waiting for the perfect, mathematically undeniable pitch, because most of the time the perfect pitch never arrives, the price recovers without you, and you call yourself disciplined when you were really just stuck.

You put a toe in.

A starting position of about thirty percent of intended size, here, around A$1.57 to A$1.59. That position acknowledges that the stock is genuinely cheap on most defensible metrics, that the existing dividend is already paying you a double-digit cash yield to sit with the position, and that the founder has put real money down at roughly this same level.

You leave the next sixty to seventy percent of your intended size unallocated. You size up only on one of two triggers.

1. The price gives you a fatter margin. Another rough quarter, another negative-eight-billion flow print, an AUD pop, or an Adani headline, and the stock visits A$1.40 or below. Then you complete the position.

2. The flow data gives you a cleaner signal. Two monthly FUM updates in a row where net flows are flat-to-positive. Or a credible CIO succession announcement that quiets the key-person risk. Then you size up regardless of the price, because the qualitative risk just got smaller, and you no longer need as much price discount to compensate for it.

You also write down, in advance and not at the moment of panic, what kills the thesis:

- Jain departs without a credible successor named at the same time.

- Two consecutive quarters of net outflows worse than US$10bn each.

- FUM drops below US$140bn.

- The 90% payout policy is reduced or suspended.

- A governance event involving QVFT.

- Material litigation or regulatory action.

If any of those happen, you exit. Not "reassess." Exit. The discipline is the pre-commitment.

The Bottom Line, Without the Adjectives

GQG Partners today is the cheapest unleveraged 77%-operating-margin business I can find on the ASX. It is run by a founder who owns seventy percent of it and is buying more of it. It pays out ninety cents on every earned dollar. It is currently being punished by the market for the *direction* of its flows while quietly compounding the *magnitude* of its profits.

It is also a stock whose bear case sits only eight percent below today's quote, whose dividend has no franking, whose moat is performance-based and demonstrably cyclical, and whose entire economic engine depends on one man's continued presence at his desk.

Both paragraphs are true. The market is currently emphasising the second one. I think the first one will matter more over ten years. I am not certain enough to bet the farm. I am certain enough to start.

A toe, not a leg. Then, with as much patience as I can manage, the willingness to do nothing for a while.

Munger again, since he said it better than anyone: *"The big money is not in the buying or the selling, but in the waiting."* The screener has done its job. The numbers have done theirs. Now the waiting starts.