Strategy Sold Bitcoin at a Loss to Pay a 12% Dividend

Between July 27 and August 2, Strategy sold 1,638 bitcoin for $104.73 million, at an average price of $63,957 per coin. The company's aggregate cost basis is $75,419. It sold roughly 15% underwater, and it was the first time Strategy has sold bitcoin since 2022.

The disclosure of where the money went is more informative than the sale itself. $52.4 million paid preferred stock dividends. $52.3 million repurchased STRC — the Variable Rate Series A Perpetual Stretch Preferred.

Michael Saylor addressed it on X on August 3: "When I say 'Never Sell Your Bitcoin,' I speak as one saver to another. I have never sold mine. Not one satoshi. Strategy is a public company, not my wallet."

Both halves of that are accurate. And the second half is precisely the thing worth looking at, because a public company has obligations that a saver does not.

The obligation

STRC carries a 12% annual dividend on roughly $10.46 billion in notional value. That is about $1.26 billion a year in cash, paid twice monthly.

Twice monthly. Not quarterly, not annually — a cash requirement that lands twenty-four times a year against an asset that produces no cash at all.

That is the whole structure in one sentence. Strategy issues fixed-income instruments with real coupons and buys an asset with no yield. The trade works on one condition: bitcoin appreciates faster than the cost of the paper financing it, and the equity trades at a premium to net asset value so new issuance is accretive rather than dilutive.

Both conditions are currently strained. Bitcoin entered August around $64,736 and spent the first week of the month in a narrow $63,000–$65,000 band — comfortably below the $75,419 basis. The coupon does not care. It arrives on the 1st and the 15th regardless of where spot is.

Strategy still holds 842,138 BTC and reported a $4.0 billion cash reserve as of August 2 for dividend and debt service coverage. It has disclosed that it may sell up to $1.25 billion of bitcoin if it needs the cash. This is not a distressed company. The sale represented about 0.5% of holdings — a rounding error against the position.

Why 0.5% is the story anyway

Because the direction reversed.

For four years the mechanism ran one way: issue equity and preferred at a premium, convert the proceeds into bitcoin, report a higher bitcoin-per-share, watch the premium justify the next issuance. Capital markets fed the treasury. The treasury never fed the capital markets.

This week the treasury paid the capital markets. Bitcoin was liquidated to service claims on bitcoin. Whatever the size, that is a different machine than the one investors were underwriting, and it is the first observed instance of it.

The buyback half compounds the signal. Spending $52.3 million to repurchase STRC is Strategy retiring its own 12% obligation — a rational, even shareholder-friendly move, and an admission that the cheapest available return on capital right now is not another coin. It is less coupon.

Selling an asset below cost to reduce the cost of the paper that bought it is what deleveraging looks like at the beginning, when it is still voluntary and still small.

The premium was always the load-bearing wall

The market-value-to-NAV premium is what made this structure work, and it is what makes it fragile.

While the equity trades above the value of the bitcoin behind it, every share issued buys more bitcoin per share than it dilutes — a genuine, self-reinforcing flywheel, and Saylor built it deliberately and early. When the premium compresses toward or below one, issuance stops being accretive, the flywheel stalls, and the fixed obligations that were funded by issuance have to come from somewhere else.

There are only two somewheres: the cash reserve, or the coins.

Strategy used the coins this time while holding $4.0 billion in reserve — which suggests the sale was a managed, structural decision rather than a liquidity scramble. Reasonable people can read that either as prudent balance-sheet management or as an early, deliberate rehearsal of a move the company would rather practice at 0.5% than perform at scale.

The imitators are the exposure

The more consequential fact is that Strategy's template got copied. A cohort of digital asset treasury companies now runs the same shape — raise paper, buy a volatile asset, market the per-share holdings — generally with smaller reserves, shorter track records, and less favorable financing terms than Strategy secured on the way up.

Strategy sold 0.5% of an enormous position while sitting on $4 billion in cash. It is the strongest participant in this category, and it is the one with the most room to be patient.

If the originator is selling below basis to make coupon payments, the imitators are running the same arithmetic with a thinner cushion, and their version of this week arrives sooner and looks worse. The question those balance sheets face is not whether bitcoin recovers. It is whether it recovers before the next dividend date.

Which is the trade nobody was pricing when the premium was 2x: not bitcoin's direction, but bitcoin's schedule versus the coupon's.

Not one satoshi from the wallet. Sixteen hundred and thirty-eight from the balance sheet.