What Strategy Inc.'s $2.1 Billion STRK Preferred Stock Filing Actually Discloses
On March 22, 2026, Strategy Inc. quietly terminated a $20.34 billion at-the-market preferred stock program. The very next day, it launched a new one — this time sized at $2.1 billion. Same instrument, same 8.00% annual dividend, fresh prospectus. On the surface, this looks like routine financial housekeeping: a program reset, nothing to see here. But buried inside that March 23 filing is one of the most candid disclosures I have read from a public company in a while. Strategy explicitly states that it plans to fund the dividends on this preferred stock primarily by selling more shares. Not from bitcoin appreciation. Not from operating income. From new capital raises. That circular structure is worth understanding in full before deciding whether STRK is an attractive yield instrument or a cleverly packaged risk transfer.
What I want to do in this post is walk through exactly how the new STRK offering works, where it sits in Strategy's capital structure, and what the honest risk profile looks like for someone buying it — because the filing lays it all out, and most of the commentary I have seen glosses over the uncomfortable parts.
The Capital Flywheel, Explained
Strategy's core financial model is straightforward once you understand the basic loop. The company raises money by selling equity or debt instruments, uses those proceeds to buy bitcoin, and then points to its growing bitcoin treasury as justification for future capital raises. Bitcoin appreciates, the treasury grows in dollar terms, the stock commands a premium, and that premium makes future share issuances more powerful. Lather, rinse, repeat.
Preferred stock sits in a specific place within this machine. Unlike common equity, preferred stock pays a fixed dividend (similar to bond interest) and has a defined liquidation preference — the amount per share that holders are entitled to recover before common shareholders receive anything if the company were wound down. Unlike bonds, however, preferred dividends can usually be deferred without triggering a legal default, which gives the issuer more flexibility during stress periods.
At-the-market (ATM) offerings allow a company to sell shares gradually into the open market at prevailing prices, rather than doing a single large block deal at a negotiated discount. Strategy has used ATM programs extensively for its common stock, and the STRK program works the same way — broker-dealers sell shares on Nasdaq throughout the day, and Strategy accumulates proceeds that it then deploys into bitcoin. The $2.1 billion ceiling on this program is simply the maximum aggregate gross proceeds the company can raise before filing a new prospectus supplement.
How STRK Is Structured
Here is the specific anatomy of this offering, drawn directly from the March 23, 2026 Form 424B5 filing on SEC EDGAR:
-
The instrument. STRK is 8.00% Series A Perpetual Strike Preferred Stock, listed on Nasdaq. "Perpetual" means it has no maturity date — Strategy is not obligated to redeem it at any fixed point in time. The 8.00% dividend is calculated on a liquidation preference of $100 per share, so each share generates $8.00 per year in dividends, paid quarterly.
-
The conversion option. STRK carries an embedded conversion feature. Each share can be converted into 0.1000 shares of Class A common stock (ticker: MSTR), implying a conversion price of $1,000 per Class A share. At Class A's closing price of $135.66 on March 20, 2026, that conversion is deep out of the money — meaning it would only become attractive if Class A stock rallied more than seven-fold from current levels. For most holders buying STRK today, the conversion feature is effectively decorative. The floor price of $119.03 per Class A share protects STRK holders from the most dilutive anti-dilution adjustments, but again, at $135.66, the margin above that floor is thin.
-
The dividend payment mechanics. This is the part the filing is unusually direct about. Strategy may pay STRK dividends in cash or, at its own election, in additional shares of Class A common stock. If it pays in shares, those shares are valued at a slight discount to market — which is effectively a forced reinvestment of your dividend back into common equity whether you want it or not. More importantly, the filing states plainly: "We currently intend to fund any dividends paid in cash on the STRK Stock primarily through additional capital raising activities, including, but not limited to, at-the-market offerings of our class A common stock and our junior preferred stock." There is no suggestion that operating free cash flow or realized bitcoin gains are the primary planned funding source. The dividend pays you today by diluting common shareholders tomorrow.
-
Outstanding shares and implied size. At the time of filing, 14,020,744 STRK shares were already outstanding. At the $75.41 sale price recorded on March 20, this new $2.1 billion program could bring the total STRK share count to approximately 41.9 million shares — roughly tripling the series. That scale of issuance matters when you start thinking about the cumulative dividend obligation: 41.9 million shares × $8.00 per year = roughly $335 million in annual preferred dividends from STRK alone, before counting the other preferred series.
The Priority Stack: Where STRK Actually Sits
Understanding an instrument means understanding what is ahead of it in the capital structure. This is the part of Strategy's preferred stack that I think deserves more attention.
As of December 31, 2025, Strategy carried approximately $8.25 billion in consolidated debt — primarily convertible notes — all of which ranks senior to STRK in both interest payments and liquidation. That debt must be serviced before STRK holders receive anything in a wind-down.
But that is not the end of the line ahead of STRK. Three separate preferred stock series also outrank it for dividends and in liquidation:
- STRF — 12,839,689 shares outstanding, paying a 10% annual dividend
- STRC — 50,246,513 shares outstanding, paying a variable rate
- STRE — 7,750,000 shares outstanding, paying a 10% annual dividend
Combined, roughly 70.8 million shares of senior preferred sit ahead of STRK. Add $8.25 billion in debt, and you have a very long line before STRK holders recover a single dollar in liquidation. The filing uses the term structural subordination to describe this, and it means exactly what it sounds like: the structure of the capital stack, not just legal agreements, creates the subordination.
To put this in concrete terms: if Strategy's bitcoin holdings were to fall sharply — and bitcoin has already swung from below $65,000 to above $120,000 in the twelve months preceding the filing date — the asset base supporting STRK's $100 liquidation preference would erode quickly. The company's entire investment thesis rests on bitcoin's long-term appreciation. STRK's recovery does too, but with far more claims in front of it.
The Numbers in Context
The prior STRK ATM program — terminated on March 22, 2026, the day before this new filing — had been sized at $20,340,632,356.64. That is not a typo. Strategy had a $20 billion preferred stock program outstanding that it chose to cancel and replace with a $2.1 billion program. Why the dramatic downsizing? The filing does not say explicitly, but the logical inference is that market appetite for STRK at current prices and the pace of actual issuances made the larger program ceiling unnecessary for near-term needs. Running a smaller, refreshed program is cleaner from a registration and disclosure standpoint.
Class A common stock (MSTR) closed at $135.66 on March 20, 2026. For reference, Class A reached an all-time high of $543.00 and an all-time low of $13.26 since the company first announced its bitcoin purchase strategy in August 2020. Those two data points — a 40x spread from low to high — tell you everything you need to know about the volatility profile of the equity layer sitting beneath STRK's claims.
As of March 19, 2026, Strategy had 325,954,147 Class A shares and 19,640,250 Class B shares outstanding. The Class B shares carry ten votes each versus one vote for Class A, meaning Michael Saylor retains effective voting control regardless of how many new shares are issued. Common shareholders do not get a vote on this strategy; they accept the dilution.
What Could Break This Thesis
Let me be direct about the specific failure modes:
-
Structural subordination in a liquidation. With $8.25 billion in debt and ~70.8 million senior preferred shares ahead of it, STRK faces a realistic scenario where a sustained bitcoin bear market depletes the asset base entirely before STRK holders see any recovery. If bitcoin were to revisit, say, the $30,000 range for an extended period, the math gets uncomfortable fast.
-
Dividend circularity breaks down. Strategy's explicit plan is to fund STRK cash dividends through new ATM equity issuances. This works as long as capital markets remain open and receptive, and as long as MSTR common stock trades at a sufficient premium to its underlying bitcoin NAV to make new issuances accretive. If that premium collapses — either because bitcoin falls sharply or because investor sentiment toward Strategy turns negative — the funding mechanism for dividends fails. The company could then elect to pay dividends in common stock instead of cash, which may not be what yield-seeking buyers of STRK are looking for.
-
Bitcoin price volatility impairs the treasury. The filing itself acknowledges that bitcoin traded below $65,000 and above $120,000 in the trailing twelve months. The entire STRK thesis depends on bitcoin's long-term value trajectory. If that trajectory reverses for long enough, the company's ability to raise new capital, service its preferred dividends, and maintain its NAV collapses in sequence.
-
IRS fast-pay stock characterization. This one is unusual and easy to overlook. The filing flags a risk that the IRS may characterize STRK as fast-pay stock under Section 7701(l) of the Treasury Regulations. If that determination is made, STRK holders could face listed-transaction reporting requirements, potential penalties, and adverse withholding tax on non-cash deemed distributions — meaning they could owe tax on income they never actually received in cash. This is a genuine and unresolved legal risk that every non-US holder in particular should think carefully about.
Conclusion
What Strategy is doing with STRK is not magic, but it is sophisticated. The instrument gives yield-hungry investors access to an 8.00% annual dividend with a bitcoin-backed collateral story, at a time when private credit and investment-grade bonds are offering meaningfully less. For a certain type of investor — one who wants fixed-income-like exposure but also believes in the long-term bitcoin narrative — STRK has genuine appeal.
But the filing is honest in a way that the marketing often is not: the dividends are planned to be funded by selling more shares, not by bitcoin profits. The liquidation claim sits behind $8.25 billion in debt and three senior preferred series. The conversion option is decorative at anything close to current prices. None of that makes STRK a bad instrument — but it makes it a very specific one, suited to investors who have already underwritten the bitcoin thesis deeply and who understand they are accepting meaningful structural risk in exchange for that 8.00% coupon. The fact that Strategy terminated a $20 billion predecessor program and replaced it with a $2.1 billion one suggests the company is calibrating issuance pace carefully. That is prudent. Whether the instrument is right for your portfolio depends entirely on how you think about the asset sitting at the bottom of all of this: a reserve of bitcoin in a company that has bet its entire capital structure on its continued appreciation.