The Model That Needs Its Own Price to Work

You've met him. The Bitcoin guy.

He's a coworker. Reasonable, with a job and a 401(k). Posts a laser-eye photo on his birthday. On one point he doesn't bend.

He's never selling his Bitcoin.

He'll die holding it. He'll borrow against it before he sells a single coin.

Michael Saylor used to talk that way too.

Saylor runs a company called Strategy. It used to be called MicroStrategy. They made business software for a quarter century. Now they buy Bitcoin. That's the business. They hold more of it than any other public company on earth.

For five years, Saylor said the same thing. Never sell. Borrow against it, hold it forever, pass it to your kids. He told an interviewer in 2022 to sell a kidney before parting with Bitcoin.

Last month, his company sold some.

Thirty-two coins. Roughly $2.5 million worth, between May 26 and May 31. A rounding error against the 843,706 Bitcoin Strategy holds. But the symbolism was bigger than the trade.

The proceeds went to a specific bill. Strategy carries about $1.5 billion in annual dividend obligations across five different series of preferred stock. One of them, called STRC, pays 11.5%. The company normally funds those payments by issuing more common shares. A sale of Bitcoin to fund a dividend … that's a tell.

The mechanism is interesting.

Strategy doesn't just hold Bitcoin. It buys Bitcoin with money it raises by selling its own stock. And that stock has traded for years at a premium to the Bitcoin sitting on the company's balance sheet. In November 2024, the premium peaked. Investors paid $3.89 for every $1 of Strategy's Bitcoin. Saylor calls the multiple "mNAV." It's the fuel.

When mNAV is above 1.0, every dollar Strategy raises by selling stock buys more than a dollar of Bitcoin for existing shareholders. That's accretive. It's the whole point of the model.

When mNAV falls below 1.0, the math flips. Selling stock dilutes existing holders. Accretion turns into shrinkage.

The premium IS the strategy.

Strategy's mNAV fell below 1.0 in early 2026. It's around 1.2x today. Not enough buffer.

So when the dividend bill came due last month, Strategy could either issue stock at a thin premium … or sell Bitcoin. They sold a small slice of Bitcoin. The CEO told analysts on the earnings call: "We will sell Bitcoin when it's advantageous to the company."

That sentence was once unthinkable.

I have no idea what Bitcoin does from here. But I know this. Strategy owns about 4% of every Bitcoin that exists. Saylor has said himself that without Strategy's weekly buying, Bitcoin might trade in the $40,000 to $50,000 range.

A buyer that big going on pause … or going the other way … reprices the asset.

The copycats are everywhere now.

More than 140 public companies hold Bitcoin on their balance sheets as of April. They collectively own about 1.16 million coins — roughly 6% of every Bitcoin in existence. Almost all of them copied Strategy's playbook. Sell stock at a premium, use the proceeds to buy Bitcoin, repeat.

The model only works while the premium holds. The premium only holds while investors keep believing. Belief depends on Bitcoin rising. And Bitcoin's price, on the margin, now depends on buyers like Strategy.

That loop runs in both directions.

Every model built on leverage has a point where the leverage stops adding power and starts removing it. The Bitcoin treasury model has a few of those points.

Strategy carries roughly $8.25 billion in outstanding convertible notes, with strikes from about $149.80 to $672.40 per share. The bondholders convert to stock if MSTR rallies past those strikes — diluting everyone else — or they can ask for cash repayment if the stock stays well below them.

On top of those sit five separate series of preferred stock paying a combined $1.5 billion in annual dividends. One of them carries an 11.5% coupon. The CEO told analysts the company has about 18 months of dividend coverage at the current run rate.

In March, Strategy doubled its capital-raise target. The plan was once $42 billion in equity and fixed income over three years. Now it's $84 billion.

That works when stock issuance is accretive. It looks different when it isn't.

And here's where it spreads.

A Japanese investment company called Metaplanet now holds about 35,102 Bitcoin. The total cost was roughly $3.78 billion at an average price near $107,600 per coin. Bitcoin trades around $67,000 today. That stack is underwater by about $1.45 billion. The company's premium still trades above 1.0 — for now.

A SPAC called Twenty One Capital came public in December, backed by Tether and SoftBank. The mandate is identical — raise capital and accumulate Bitcoin.

A small medical-devices company called Semler Scientific spent 2024 and 2025 buying Bitcoin until its premium dropped below 1.0. The asset manager Strive then acquired the company partly for the Bitcoin sitting on its balance sheet.

"BTC Yield is financial gibberish." — Jim Chanos

Chanos has been wrong on plenty of trades. But he's right about the metric. Bitcoin acquired per share issued only makes sense when the share trades at a premium to the Bitcoin behind it.

Strip the premium out and the math is much simpler.

Meanwhile, the largest corporate buyer of Bitcoin has now sold some of it.

The arithmetic underneath that sentence is bigger than most readers realize. Strategy owns about 4% of every Bitcoin that has ever been mined. An estimated 20% of all Bitcoin issued is lost forever — keys gone, wallets forgotten. So the company's share of the available float runs well above 4%.

Saylor has said publicly that without Strategy's consistent weekly buying, Bitcoin might trade in the $40,000 to $50,000 range. That's an admission worth holding on to.

Corporate treasuries as a group added about 62,000 Bitcoin in the first quarter of 2026. Most of those buyers used the same playbook. When that playbook stops working for one of them, it stops working for the rest.

And when the marginal corporate buyer turns into a marginal corporate seller, the implied price floor disappears with it.

That's the loop. It's already running in reverse, in tiny doses.

Three more things worth keeping track of…

Subprime auto loans are at a 32-year delinquency record. Fitch puts the 60-day-plus delinquency rate on subprime auto ABS at 6.90% as of January, the worst reading in tracking that goes back to 1993. Cox Automotive estimates 1.73 million vehicles were repossessed last year, the highest annual total since 2009. The pain is concentrated entirely in subprime — prime auto delinquencies sit near 0.4%. Lower-income consumers are running out of room.

Office CMBS delinquencies hit a record 12.34% in January. That's the worst reading Trepp has on file going back to 2000. Roughly $25 billion in CMBS loans now sit past maturity without payoff, liquidation, or formal extension. About $100 billion in commercial real estate–backed CMBS matures in 2026, and Morningstar DBRS expects more than half of those loans to default. The "extend and pretend" era is winding down.

The New York Fed's Empire State Manufacturing Index collapsed in June. The headline reading dropped from +19.6 in May — a four-year high — to +5.7. The New Orders subindex fell 19.2 points to +3.5, the slowest pace of the year. Shipments fell more than 10 points. It's one region and one survey. But the speed of the reversal matters. We'll see.

There's a phrase I keep going back to. Financial whoopee.

Bernard Baruch said it in the summer of 1929. He was being offered a piece of a new stock — an investment trust called Shenandoah Corporation — and he turned it down with that line.

Shenandoah was launched on July 26, 1929 by another investment trust called Goldman Sachs Trading Corporation. The Trading Corporation itself was only seven months old. Goldman Sachs incorporated it in December 1928 and sold most of the shares to the public at $104. Investors loved it. The premium ran all year.

When Shenandoah came public, demand was so heavy the offering was oversubscribed seven times. Stock was issued at $17.50. It opened at $30. It closed the first day at $36.

Twenty-five days later, Shenandoah sponsored a third trust called Blue Ridge Corporation. The structure was the same. Blue Ridge held shares of things Shenandoah liked. Shenandoah held shares of things Goldman Sachs Trading Corporation liked. Each trust collected the gains of the one underneath it. And each trust traded at a premium to the value of what it owned.

"The principle of leverage is the same for an investment trust as in the game of crack-the-whip. By the application of well-known physical laws, a modest movement near the point of origin is translated into a major jolt on the extreme periphery." — John Kenneth Galbraith

In August 1929, an estimated 60% of all new securities issued in the United States were investment trusts. In September alone, more than $600 million in investment-trust securities came to market. The whole model required investors to keep paying premiums for shares of trusts that held shares of trusts.

The crash came in October. The premiums dropped. Forced selling began.

Goldman Sachs Trading Corporation peaked at an adjusted share price of $326. By mid-1932, it traded at $1.75. A 99% loss.

Shenandoah fell from $39 to about 50 cents. Blue Ridge fell from $29 to 63 cents.

The mechanism worked beautifully in one direction. It worked just as efficiently in reverse.

I don't have a date for the next one. I just keep coming back to the phrase. We'll see.