When a Bitcoin Treasury Company Turns From Buyer to Forced Seller

For most of 2026, the steadiest buyer of bitcoin was not an ETF. It was a bitcoin treasury company, a firm that had told its shareholders it would never sell.

Then Citi did the arithmetic in public. On July 1 the bank cut its twelve-month bitcoin target to $82,000 from $112,000, marked its assumed net ETF inflows down from $10 billion to zero, and pointed at a source of selling that few people had priced in: the treasury companies themselves. The same note trimmed its ether target to $2,240.

The structure that makes such a company a forced buyer on the way up can make it a forced seller on the way down. Not because management panics. Because the math changes sign.

This is a Kodex walkthrough built as a conversation. Lucia asks what a careful reader is already asking. Lilith answers: twenty years spent reading systems, she watches where selling pressure settles once the drama on the screen has moved on. What the two of them take apart is a single number, mNAV, and why it decides whether a treasury company is bitcoin's backstop or its overhang.

Lucia already has the June ETF figures open. "About four billion dollars left the spot bitcoin ETFs last month," she says. "Worst month on record. That is the story. The sellers are the funds."

Lilith does not look at the ETF column. "That is one pipe," she says. "Citi just put a flashlight on the second one."

What a bitcoin treasury company actually holds

"Start with what the company is," Lilith says. "A bitcoin treasury company, sometimes called a DAT, holds bitcoin as its main reserve and pays for more of it by issuing its own stock. The share is the straw. The bitcoin is the drink."

Lucia frowns. "So it is a bitcoin fund with a ticker."

"Close enough to be dangerous. The pitch was better than a plain fund. Buy the stock and you got bitcoin exposure inside a public company that could borrow, issue shares, and accumulate far more coins than your own cash ever would. Bitcoin with a public-market wrapper your brokerage already sold you. That is why the model spread as fast as it did."

She taps the desk. "A fund holds what it holds and tracks it. A treasury company can raise fresh money and add to the pile, or it can be pushed into shrinking it. Which of those happens comes down to one ratio, and the ratio has a name."

She writes it out. mNAV: the company's market value set against the bitcoin it holds. Market capitalization over coins, near enough for what matters here. Separating a company's market value from the assets sitting underneath it is its own skill, and the What Is Crypto Market Cap walkthrough is the place to slow down if that split is new.

"When the market pays more for the shares than the bitcoin under them is worth, mNAV sits above 1.0," Lilith says. "When it pays less, mNAV drops below 1.0. Through the first half of 2026, one company after another slipped under."

"And that matters because," Lucia says.

"Because 1.0 is the line where the machine runs forward or in reverse."

The flywheel only spins above 1.0

"Above 1.0, the design runs in the holder's favor," Lilith says. "The company sells one new share for more dollars than the bitcoin that share will buy. It pockets the difference, buys bitcoin, and every existing shareholder walks away with more bitcoin per share than they had that morning. Issue, buy, repeat."

"Dilution that makes the holders richer," Lucia says slowly. "That should not be possible."

"Put a number on it. Say the stock trades at a forty percent premium to the coins inside it. The company sells a hundred and forty million dollars of new stock, buys a hundred and forty million dollars of bitcoin, and the bitcoin standing behind every old share goes up, not down. Do that quarter after quarter and the holding per share compounds."

Lucia nods. "So the premium is not vanity. It is the fuel."

"It is the whole engine. And that is the sentence nobody says at the top. The flywheel does not run on belief in bitcoin. It runs on the share price staying above the value of the coins the company already owns."

The premium is the fuel. Not the conviction.

"Strategy wrote this playbook," Lilith says. "For years it traded at a fat premium, issued stock and preferred shares, and bought bitcoin at prices that grew its holdings per share. On June 27 its mNAV fell below 1.0 for the first time, to about 0.99. The engine did not blow up. It stalled."

Lucia looks up. "Stalled how? It still owns all that bitcoin."

"It owns every coin. What it lost is the ability to add cheaply. A machine built to keep adding does not stand still gracefully once it can no longer add."

What happens when mNAV falls below 1.0?

"Walk me through the reverse," Lucia says. "The premium is gone, the stock trades below the bitcoin it owns. So what? It is a paper number. Nobody has to do anything."

"Someone almost always does," Lilith says. "Sit in a shareholder's seat. You hold a stock worth ninety cents for every dollar of bitcoin inside it. You start asking why the company does not just buy back its own shares and hand you that gap. It is the obvious move. Retire a share at that discount and the holders who stay get a dollar of bitcoin backing for eighty or ninety cents spent. On paper it reads like free money."

"So they buy back stock."

"With what? A company that spent every dollar it ever raised on bitcoin has no cash parked for a buyback. So the pressure runs to the one asset it does have. It sells bitcoin to fund the buyback, or to cover whatever else is coming due."

The buyer becomes the seller. Same company, same structure, opposite direction.

"That is the flip," Lilith says. "Not a loss of faith. A change of sign. Above the line, issuing stock pulls bitcoin in. Below the line, defending the stock pushes bitcoin out."

Lucia sits with it. "And the moment it sells, that is not paper anymore."

"No. That is real coins landing on the order book, from a holder the market had written down as permanent."

The second sell pipe Citi just named

"And this second pipe is separate from the ETF outflows I was looking at," Lucia says.

"Separate, and invisible next to them. Redemptions and treasury selling are two pipes feeding the same order book. When you read a spot bitcoin ETF outflow, you are watching one specific mechanism: an investor sells the fund, an authorized participant hands back shares, the fund delivers or sells the coin. It is measurable. It prints daily."

"And the treasury pipe?"

"Does not show up in that data at all. A treasury company selling from its reserve is not an ETF redemption. It is a corporate decision that lands on the same exchanges, prices into the same book, and carries none of the same visibility."

So how do the two pipes differ? Lilith lays them side by side.

Read only the first column and you will believe you have measured the selling. You have measured half of it.

That gap is what moved Citi. The bank did not only remove a buyer when it cut assumed ETF inflows from ten billion dollars to zero. It added a supplier on the other side by naming treasury companies as a fresh source of selling. Fewer coins wanted, more coins potentially for sale, and the target came down.

It is not hypothetical. K Wave Media told the SEC on June 30 that it had sold its entire bitcoin position, about $64.2 million, all of it. A holder the market had filed under permanent turned out to be temporary.

"There is a nastier wrinkle," Lilith says. "The treasury pipe does not open at random. It opens when bitcoin is already falling, because that is what drags mNAV below 1.0 in the first place. So the selling this structure produces arrives correlated with the very weakness that triggered it. A redemption can hit on a flat, quiet tape. Forced treasury selling shows up precisely when the book is already thin and the bids are already backing away."

Lucia sees it. "The second pipe opens at the worst possible moment."

"By design, not by accident. That is what makes it worth watching before it starts."

When a discount does not force a sale

Lucia leans back. "So every company trading under 1.0 is dumping bitcoin as we speak."

"No," Lilith says. "And that is the line careful reading has to hold, or you will scare yourself out of every position you own."

mNAV below 1.0 is the condition. It is not the trigger.

"A treasury company with cash in the bank, no debt coming due, and a board that can wait will sit under 1.0 for months and do nothing. Buy nothing, sell nothing, ride it out. The forced sale needs two things in the same room: the discount, and a bill that cannot be paid any other way. Debt maturing. A margin call on borrowed money. A buyback the company already promised out loud."

The discount is the setup. The cash crunch is the trigger.

"One without the other is noise," Lilith says. "A cheap stock is not a seller. A cheap stock with a bond due next quarter and an empty account is a seller. The skill is telling those two apart before the market does."

"A discount alone gets read as dumping," Lucia says.

"A discount alone means a company is uncomfortable. It does not mean it is cornered. Those two get treated as the same thing, and the difference between them is the entire trade."

How to read a treasury company's sell pressure

"Then tell me what I actually watch," Lucia says.

"Three things, in order," Lilith says.

Where mNAV sits, and which way it is moving. Above 1.0 and climbing, the company is a buyer and a bid under the market. Below 1.0 and sliding, it moves onto the watch list.

Whether it holds cash or spent it all on coins. A treasury with a cash cushion controls its own timing. A treasury that is all bitcoin and no cash has handed its timing to the market.

When its debt comes due. A quiet balance sheet with nothing maturing can outlast almost any discount. A wall of debt in the next two quarters turns a paper problem into a calendar.

"Put those together and you are not predicting the price," Lilith says. "You are reading which holders can be pushed, and when. A company at 1.4 with cash is structurally a buyer, whatever it says on a bad day. A company at 0.8 with a bond due and an empty account is a candidate to sell, whatever its founder posts at midnight."

Lucia looks back at her ETF column. The four-billion figure has not changed. Her reading of it has.

"That number tells me what left through one door," she says.

"It tells you nothing about who is standing at the other," Lilith says. "That is the Survival Framework move. Read the structure that decides who has to act, not the headline that tells you the selling is finished."

The print measures one pipe.

Now you know there are two.