Strategy's Second Cash Pot, and the $1.28bn It Has Spent on STRC

When I wrote about Strategy's Digital Credit Capital Framework on 30 June, the USD Reserve stood at $2.55 billion and there was one dollar pot on the diagram. Strategy's 8-K for the week to 27 September gives two numbers instead: a USD Reserve of $5.02 billion and a second pot, USD Cash, at $1.00 billion, that didn't exist in June. In the same period, Strategy bought back $1.28 billion of its own STRC. STRC closed on 30 September at $99.35. That is close to par, but it is still not $100, and I think the gap between those two numbers is the most important thing in this story.

So this is a follow-up to Strategy Transforms and Working Backwards. It covers what the new pot is, where the buyback money has gone, why STRC still can't do its original job, and how the proposal to pay dividends every calendar day fits in. Every Strategy figure below comes from its own weekly 8-K filings; the averages and totals are my arithmetic on those figures, and I say so where it matters.

Two pots, two jobs

On 24 August Strategy announced USD Cash as a new part of the framework. The 8-K describes it as a separately designated pool of dollar liquidity that the company can keep for future use, including buying Bitcoin, paying declared preferred dividends and interest, buying back MSTR or preferred stock, repaying convertible notes, or topping up the USD Reserve. It also says the USD Reserve policy is unchanged. The Reserve is still the ring-fenced one, and USD Cash is the flexible one.

The Reserve roughly doubled between the framework announcement and late August, mostly through sales of MSTR common stock, and has been flat since. Strategy's own filings caution that the balances include cash from share sales that hadn't settled yet, so the real settled figure can lag the headline slightly. Strategy's own cover figures imply an annual bill of roughly $1.76 billion for preferred dividends and interest: in June it put $2.55 billion at 17.4 months, and on 30 July it put $3.75 billion at "more than 2.1 years", and both work out at about $1.76 billion a year. Its wording is "existing preferred dividend payments and interest obligations", which I read as every series paid from that one pot, so STRF, STRK, STRD and the euro STRE as well as STRC, plus interest on the debt. On that bill, $5.02 billion is about 34 months of cover, against the board's 12-month floor. That division is my arithmetic, and it assumes the bill hasn't changed; the buybacks will have trimmed it a little. The Reserve has also started to be used for its stated purpose: $57.4 million of it paid dividends and interest in the week to 20 September, and $22.1 million the week after.

I think the second pot answers a question the June framework left open. The Reserve is deliberately hard to touch, which is what makes it reassuring, but it also means Strategy had no big pile of cash it could move quickly. USD Cash is that pile. The 8-K says the extra flexibility is meant to let management respond faster to dislocations in the markets for Bitcoin or Strategy's securities, and as the next section shows, it was being used for exactly that within two weeks.

The picture, updated

Here is the diagram from Strategy Transforms with the new pot added. Cyan is USD Cash. Notice that the buyback arrow no longer has to come from Bitcoin sales: since early September it has come largely from the new pot.

Where the $1.28 billion went

Nothing happened on the buyback front for the three weeks after the 29 June announcement. Strategy's 8-Ks for the weeks to 5, 12 and 19 July all say it bought no shares under its repurchase programmes. Then, in the week to 26 July, it started, and it bought STRC in every one of the ten weeks up to 27 September:

*Spend divided by shares, my calculation from Strategy's weekly 8-Ks. The total matches the programme's own numbers: $2.0 billion authorised on 8 September, $723.5 million left at 27 September.

Two things stand out to me. The first is the price column. The average paid has climbed from $86.53 to $98.86 in ten weeks, which is the recovery in STRC's price showing up in Strategy's own receipts. The second is the pace. Strategy raised the programme from $1 billion to $2 billion on 8 September, and it has been spending about $150 million a week. At that rate the $723.5 million left lasts roughly five weeks, again my arithmetic, so I expect either a pause or another increase before the end of November.

On the dividend side, 13.27 million shares at the $12 a year that 12% pays on $100 is about $159 million a year of dividend bill that Strategy no longer has to meet, assuming the shares are retired. That is a real saving, though against $1.28 billion spent to get it, it is a long payback and not the point of the exercise. The point is the price.

The funding has changed along the way, which is where the new pot comes in. In July and early August the buybacks were part-paid by selling Bitcoin: four sales between 29 June and 9 August, 6,916 coins for about $429 million in total, with the last sale in the week to 9 August. Since then Strategy has sold no Bitcoin. From the week to 7 September the 8-Ks say the STRC repurchases were paid from USD Cash: $176.3 million, $139.3 million, $174.0 million and $48.1 million, which is $537.7 million in four weeks. USD Cash fell from $1.61 billion on 30 August to $1.00 billion on 27 September, so the buybacks account for most of the drop and a $75.7 million Bitcoin purchase for the rest.

The stack has also been rebuilt. Strategy held 847,363 BTC on 28 June, fell to 840,447 after the sales, and held 847,666 BTC on 27 September, 7,218 coins bought since the low and 303 more than before the first sale. Those purchases came from MSTR stock sales and, in one week, USD Cash. None came from STRC.

What the money bought: a curve that flattens

You'd expect each dollar spent to move the price less as STRC gets closer to par, and Strategy's filings let me test that. I lined up each week's buyback spend against STRC's closing price on the last trading day of that week (Yahoo Finance), starting from $85.29 on 17 July, the last close before any buying. The chart plots cumulative spend along the bottom against the price on the side.

Weeks end on Sunday; the price used is the last close of each week. Price changes are measured from $85.29 on 17 July, and the last-three-weeks row from the 7 September close of $97.75. All of it is my arithmetic on Strategy's 8-Ks and Yahoo Finance closes.

The shape is what I expected, and the size of the flattening is bigger than I expected. The first $483 million took the gap to par from about $14.70 to $3.82. The next $793 million took it down by only another $2.36, which makes each dollar about seven and a half times less effective. In the last three weeks Strategy spent $465 million and the price rose by about 79 cents, and in the final two of those weeks it didn't rise at all. The closer STRC gets to $100, the more it costs to move it a little further.

There are limits to what this proves. STRC had already climbed from a closing low of $74.57 on 26 June to $85.29 by 17 July with no buying at all, after the rate went to 12% and the framework was announced, so a good part of the recovery wasn't down to the buybacks. Bitcoin, the 12% rate and general sentiment moved the price too, and ten weekly points is a small sample. What the chart does show is the price of staying where it is: $1.28 billion so far, and a running cost of around $150 million a week to hold a price that is no longer rising. What I can't tell from this is how much of the price would drain away if the buying stopped. That's the test I'm waiting to see.

Why STRC still can't fund Bitcoin

This is the part I keep coming back to. STRC was built to run one way: when it trades at or above $100, Strategy sells new shares into the market and uses the money to buy Bitcoin. In every weekly 8-K since the framework was announced, the STRC line under at-the-market sales is zero, and the $17.5 billion of capacity has not been touched since the end of June. Strategy is not issuing STRC. It is buying it back.

At $99.35 STRC is within 1% of $100, so it's tempting to say the job is done. It isn't, for two reasons. Selling new STRC at $99 to buy Bitcoin means issuing something Strategy may later have to redeem at $100, and the whole design depends on the price sitting at par or above, not hovering under it. And Strategy itself has been clear that it's not there yet. Its 30 September 8-K says management will recommend holding the rate at 12.00% until STRC has demonstrated "sustained, healthy trading near $100 per share", which is a higher bar than touching $99.50 for a day. STRC touched $71.25 intraday on 26 June, so the climb has been large, and much of it has been paid for.

I'd put it this way. Until STRC crosses $100 and stays there, the preferred is a cost centre. Common stock sales are paying for the Bitcoin, and the dollar pots are paying for the price support.

The 365-day proposal, in that light

I wrote about the proposal to pay dividends every calendar day when it was announced on 25 September, in STRC Goes Daily (Again). The 30 September 8-K adds the mechanics. Shareholders vote on 28 October. If it passes and the amended terms are filed by 1 November, the semi-monthly STRC dividend that would have been paid on 15 November is replaced by daily dividends for November, which Strategy intends to declare on or about 15 October, still at 12.00% a year. If either step fails, the 15 November dividend is paid as normal. The 31 October payment goes ahead either way.

It isn't only STRC. The vote covers all four of Strategy's listed preferreds: STRF, STRK and STRD are being asked to move from quarterly payments to daily ones, and the 25 September 8-K names those four, not the euro-denominated STRE. STRC would switch first, with its first daily payment on 2 November. STRF, STRK and STRD stay quarterly through their payments at the end of December, then pay their first daily dividends on 4 January 2027, for the record dates of 1, 2 and 3 January. Rates and annual totals stay the same on all four. The jump is much bigger for the three quarterly ones, from four payments a year to 365, and I covered the full comparison in STRC Goes Daily (Again).

Strategy describes the change as a way to smooth the stock's behaviour and make it easier for institutions to hold. Put it next to the buyback table and I read it as the same goal pursued with a different tool. The buybacks smooth the price by spending money: $1.28 billion so far, from pots Strategy has to keep refilling. Daily dividends aim at the same steadiness by changing the calendar and costing nothing. They don't change the rate or the annual total, so they can't pull STRC through $100 on their own, but they could reduce how much cash needs to be spent holding it near there. I can't say how much, and nor can Strategy, because nothing like it has been tried.

My view, and what I'm watching

The two-pot structure is a sensible fix for a real gap. The Reserve gives holders a number they can rely on, the 34 months of cover. USD Cash gives Strategy room to act without touching it. What I can't tell yet is how it is refilled. USD Cash fell by $610 million in four weeks, and it grows when MSTR stock is sold, so the useful signal is how often Strategy is selling common stock and at what price. A pot that is topped up by selling MSTR to buy back STRC is a transfer between shareholder groups, and holders of STRC should know that is what is being paid for.

For the next few weeks I'll be watching three things: whether STRC settles above $100 without the weekly buying, how the USD Cash balance moves once the buyback programme is topped up or runs out, and the 28 October vote. If the price holds with less spending, the framework has worked. If it only holds while Strategy is spending $150 million a week, it hasn't yet.

Where to look next

The STRC hub has the live price and effective yield, Strategy Transforms covers the June framework this builds on, and Working Backwards explains the buyback mechanism in detail.

I track STRC, SATA, BMNP and CHAD daily and hold positions in Strategy's parent company (MSTR) and in Strive's (ASST). The Strategy figures are from its weekly 8-Ks to 27 September and its 30 September filing; the STRC closing price is from Yahoo Finance. The 365-day plan is a proposal pending a shareholder vote. Not financial advice.

This article is for educational purposes only and does not constitute financial advice. STRC is a speculative investment tied to Strategy's Bitcoin holdings and capital structure. All rates, prices and figures cited are subject to change. Always consult a qualified financial adviser before making investment decisions.