The Premium on a Bitcoin Treasury Stock Is a Bet the Flywheel Still Works

Two years ago the flagship of the bitcoin treasury trade was priced like a rocket ship. In late 2024, MicroStrategy — now called Strategy — traded at roughly 3.9 times the value of the bitcoin it owned. Investors were paying $3.90 for every dollar of bitcoin sitting on the balance sheet. Everyone knew it, and most people thought it was fine.

Then the rocket came down. By the end of December 2025 the same stock traded at 0.91 times its holdings — a discount to the assets it actually owns. By June 2026 its market value had fallen below the market value of its bitcoin entirely, even as the company sat on roughly 845,000 BTC worth about $66 billion.

Bitcoin fell during that stretch. But the price of bitcoin was the smaller half of the story. The larger half was a single number that most retail investors never look at: the premium. Understanding why that number moves up and down — above and below the value of the holdings — is the whole difference between understanding a treasury stock and just chasing bitcoin's chart.

What the premium actually measures

The metric is called the modified net asset value multiple, or mNAV. The arithmetic is simple: take the company's market capitalization and divide it by the value of the bitcoin it holds. Above one is a premium. Below one is a discount.

That ratio is the entire business model. Bitcoin treasury companies do one thing: buy bitcoin, wrap it in a public company, and — the crucial part — raise new money to buy more bitcoin. The premium is what makes the engine run.

Here is the mechanism, in the cleanest possible form. Imagine the stock trades at 2 times its bitcoin per share. The company issues new shares at that inflated price, takes the cash, and buys bitcoin at 1 times its value. Even though the share count goes up, each existing share now backs more bitcoin than before. The new issuance accretes value instead of diluting it.

Galaxy Digital, the asset manager, framed it exactly this way: treasury companies trade at premiums because of their unique access to public capital markets, which lets them function as high-beta bitcoin proxies. The more shares they sell above net asset value, the more bitcoin per share they compound. That self-reinforcing loop is the "flywheel."

This is why the premium reaches absurd heights in bull runs. Metaplanet, the Japanese treasury company, traded at a 237% premium in July 2025. MSTR peaked near 3.9x in late 2024. Buyers weren't irrational — they were buying the engine, not the bitcoin. Every dollar of premium is a dollar of potential leverage the company can spend to outgrow holding bitcoin outright.

The flywheel runs on a premium — and dies without one

Here is the part most people miss, and it is the whole trade in reverse.

The flywheel only works while the stock trades above its holdings. The instant the multiple falls to one — or below — the exact same capital raise stops compounding and starts destroying value. Issuing shares at a discount to what they back means each new share dilutes the bitcoin per share of everyone already in. The engine that made the stock special has inverted into a liability.

That inversion is what happened to the entire sector in 2025–2026. Strategy's multiple fell from roughly 3.9x to 0.91x as bitcoin slid from its October 2025 peak near $126,000. By September 2025, K33 Research counted one in four public treasury companies trading below their own holdings; by January 2026 the figure reached roughly 40% of the top 100. Metaplanet's 237% premium collapsed into a 10% discount.

The reason the fall compounds is fixed cost. Strategy carries about $800 million a year in preferred-stock dividends — cash that has to be paid regardless of bitcoin's price. When the premium is gone, the old answer to "how do we pay for this" — raise shares at 2x and buy more bitcoin — no longer works. The new answer is to sell bitcoin. In mid-2026, after formalizing a "Digital Credit Capital Framework," Strategy sold about 6,948 BTC for roughly $432 million to fund dividends and build its cash reserve. The "never sell" pledge quietly ended.

None of this is insolvency. Strategy's debt load is light relative to its assets — around $8.2 billion in convertible notes against roughly $59 billion of bitcoin, a loan-to-value ratio near 14%. The convertibles don't mature until 2028–2032. The company can survive. The point is more subtle: a discount means the compounding engine has stopped, and the stock's only reason to exist — growing per-share bitcoin faster than just holding bitcoin — is suspended until the premium returns.

Why some names are permanently on the discount side

The premium is not a prize any treasury can win. It is reserved for the scaled, credible leader with an unbroken record of accretive accumulation — which in practice means mostly Strategy. Market leadership, investor trust, and years of buying are what let a company issue stock above net asset value. That is why Galaxy's numbers were so lopsided at the peak: Metaplanet at a 384% premium, The Blockchain Group at 217%, while most of the field got nothing.

The rest of the sector trades at a discount not because it is cheap but because it has no flywheel to sell. The bitcoin miners are the clearest example. MARA and Riot hold large treasuries, yet their stocks trade below those holdings and are priced like operating companies, not bitcoin proxies — Riot even rallied sharply in 2026 as investors re-rated it as a data-center developer rather than a bitcoin holder. Miners have electricity bills, hardware costs, and constant dilution; their equity is priced on earnings, not on the value of the coins in the vault.

A discount tells you nothing by itself. It only means the market does not believe this company can compound per-share bitcoin. Sometimes that is a mispricing waiting to be corrected. Often it just means the engine was never there.

The premium is the leverage — in both directions

There is an older version of this same story. Years ago, the Grayscale Bitcoin Trust traded at a 40% premium to its bitcoin, then slid to a 50% discount. Premiums flip. They flip fastest when the thing being leveraged stops going up.

That is the honest way to read any treasury stock. The premium is not a bonus on top of the bitcoin — it is the leverage. When Bitcoin rises, a premium stock climbs faster than the coin, because the flywheel lets it buy more. When bitcoin stalls or falls, the same premium is the first thing to disappear, and the stock falls faster than the coin. In 2026, Strategy's shareholders ate both bitcoin's decline and the vanishing premium — falling about 39% year to date versus roughly 27% for the spot bitcoin ETF alone, a gap that had nothing to do with leverage and everything to do with the multiple compressing toward one.

Bitcoin was back near $77,000 in September 2026 after a low near $58,000 — still far below its all-time high above $125,000. A recovery in price can revive a premium. It does not happen automatically.

So the question to ask of any treasury stock is not "premium or discount?" It is narrower and more useful: can this company still raise money above the value of its bits, and buy more without shrinking what each share already owns? Pay a premium and you are paying for that engine — knowing the engine is also the part that can break. Buy a discount and you are betting some catalyst will turn it back on. The multiple is not a valuation detail. It is the entire investment.