Bitcoin Magazine

Should Bitcoin Companies Build USD Reserves? Understanding The Truth

Strategy’s U.S. dollar reserve has reached $4.65 billion, up from $3.75 billion two weeks earlier. It has also sold almost 7,000 BTC since late June 2026. 

Many are wondering why a company built around accumulating Bitcoin would choose to hold billions of dollars in fiat. More importantly, should other Bitcoin businesses do the same? 

Strategy holds cash because it’s in a very unique position 

Strategy increasingly operates as an issuer of Digital Credit: preferred securities backed economically by an enormous Bitcoin balance sheet. These instruments create fixed dollar dividends obligations.

Bitcoin produces no cash flow. Strategy’s software business produces far too little cash to cover its capital structure. 

Traditional credit analysis compounds the problem. S&P assigned Strategy a B- rating in October 2025, citing its Bitcoin concentration, weak dollar liquidity, and very weak risk-adjusted capital. Under S&P’s methodology, Bitcoin is effectively excluded from the capital base used for this analysis because of its market risk. 

In our coverage of the S&P rating, we specifically mentioned that a cash reserve, amongst other things, was worth exploring to improve credit ratings. 

Strategy therefore holds dollars to support its credit issuance. That is literally the whole reason. 

More dollar liquidity can improve the perceived safety of its preferred securities, broaden investor demand, and potentially lower its cost of capital—in the eyes of credit ratings agencies. 

The cash still carries an economic cost. Excess capital should produce a return. A conventional company can reinvest it, repurchase shares, or distribute it. A Bitcoin company can buy more Bitcoin. Every dollar held in cash replaces potential positive returns with guaranteed negative real returns.

Strategy accepts that cost because its business model depends on issuing more credit. Three unusual conditions exist at once: Bitcoin dominates its balance sheet, rating agencies heavily penalize that Bitcoin exposure, and management intends to keep issuing Digital Credit.

All three conditions are pretty unique individually and it is exactly the combination of all three that creates the situation where they need to hold cash. For instance, if Strategy did not want to issue credit, then it wouldn’t need the cash. 

The economic consequence of cash reserves 

The math creates some glaring problems with cash reserves.

Suppose Strategy issues $100 of preferred stock carrying a 10% annual dividend and holds three years of dividend coverage in cash. It must reserve $30 and can deploy only $70 into Bitcoin.

The preferred still costs $10 per year. The $70 invested into Bitcoin must therefore generate:

$10 ÷ $70 = 14.29%

A stated 10% cost of capital becomes a 14.29% hurdle rate on the capital actually deployed. The reserve raises the required return by 42.9%. Interest earned on the cash reduces the hurdle somewhat, but the structural drag remains. 

The true hurdle is actually higher, however, because BTC’s volatility means it will heavily underperform the hurdle rate in some years, and these years still require the dividends to be paid (here I am assuming that dividends are not skipped). So aside from the cash drag, there is also a volatility drag imposed by attempting to amplify a volatile asset. This risk must be compensated for by adjusting the hurdle rate higher. 

The larger the required reserve, the less of every new dollar reaches Bitcoin. If Bitcoin appreciation fails to exceed this higher hurdle over time, common shareholders bear the cost.

However, cash is far from useless. Cash creates useful optionality. It can cover dividends and interest during Bitcoin drawdowns, reducing the risk of forced Bitcoin sales. It can also support opportunistic repurchases of securities when they trade below their stated value.

Strategy recently did exactly that. In late July, it paid $25 million for $28.89 million of STRC stated value, a 13.47% discount. It later used $108.6 million from Bitcoin sales to retire another 1.15 million STRC shares. Buying preferred stock below par removes more senior claims and future dividend obligations than the cash spent. It is also accretive to Net Bitcoin Per Share.  

Should Bitcoin companies accumulate cash or bitcoin? 

For most Bitcoin companies, cash needs should be tied to the operating business rather than to an arbitrary reserve target—consider that Strategy literally does not know how much reserves it needs to get a better rating or for more credit investors to become interested in STRC. 

A cash-flowing company usually has a good understanding of its cash outlay. It should hold enough dollars to cover payroll, taxes, debt service, vendor payments, near-term capital expenditures, and a reasonable buffer for volatility in operating cash flow.

The right reserve depends on the stability of those cash flows. A profitable business with recurring revenue, low fixed costs, and predictable expenses can operate with a smaller buffer. A cyclical or capital-intensive business needs more. The reserve should rise because the business requires liquidity, not because management simply wants a large cash balance.

Once operating needs and a prudent liquidity buffer is covered, additional cash needs a specific economic purpose. Otherwise it dilutes returns by generating a large opportunity cost. For any company, excess capital should compete directly against the company’s hurdle rates, repurchasing undervalued shares, reducing expensive liabilities, or investing in projects that can earn a higher return. 

In conclusion, Strategy is a very, very rare case. Its cash reserve exists only because it is building a large credit issuance business on top of a Bitcoin balance sheet while credit ratings agencies impose significant institutional inertia which treats legitimate, liquid assets as zero value. Companies without that liability structure—which is basically all other companies—have far less reason to accumulate dollars beyond their working capital buffer.

Disclaimer: This content was prepared on behalf of Bitcoin For Corporations for informational purposes only. It reflects the author’s own analysis and opinion and should not be relied upon as investment advice. Nothing in this article constitutes an offer, invitation, or solicitation to purchase, sell, or subscribe for any security or financial product.

This post Should Bitcoin Companies Build USD Reserves? Understanding The Truth first appeared on Bitcoin Magazine and is written by Allard Peng.

Bitcoin Magazine

Corporation’s Approach to the BIP-110 Soft Fork

BIP-110 is approaching its first consequential activation boundary. The proposal enters mandatory signaling at block 961,632, currently projected around August 9, 2026. It locks in no later than block 963,648, roughly in late August, and activates its new transaction rules at block 965,664, currently projected for early September. BIP-110 uses a 55% signaling threshold and would enforce its restrictions for 52,416 blocks, approximately one year.

Bitcoin resolves consensus changes through coordination among miners, users, and nodes (note that anyone can be any combination of these three things). Miners choose which valid chain to extend. Users decide which chain’s coins, deposits, and payments they recognize. Nodes independently choose which rules they enforce. Durable consensus emerges whenever these groups converge on the same chain.

BIP-110 restricts large data pushes, oversized output scripts, undefined witness versions, Taproot annexes, deep Taproot control blocks, OP_SUCCESS opcodes, and certain Tapscript conditionals. It grandfathers UTXOs created before activation, while standard monetary uses remain compatible with its rules.

Most corporations don’t have to do anything 

For most corporations, BIP-110 requires no action. Today, the typical corporate Bitcoin utility is as a store of value, as a long-duration treasury reserve asset. This use case is basically unaffected by the transaction features targeted by BIP-110.

Corporations using Bitcoin for payments also face limited direct impact. Standard on-chain payments remain compatible (see below for specifics), while ordinary Lightning payments occur off-chain. A chain split can still affect Lightning channel monitoring, force-close behavior, and the chain source that a Lightning node treats as authoritative. However, even corporations using Bitcoin for payments normally use a third party provider like Square, so all of this abstracted away to be a non-issue. 

A corporation that runs its own full node has a direct choice. Every user retains the right to run the Bitcoin implementation that matches its needs. A corporation that supports BIP-110 should therefore switch over to running BIP-110. All other node-running corporations can simply do nothing. 

A BIP-110 node enforces tighter rules. During mandatory signaling, it rejects blocks that fail to signal bit 4. After activation, it also rejects blocks containing transactions that violate BIP-110. A non-BIP-110 node accepts BIP-110-compliant blocks as well as blocks that remain valid under the existing rules. Among all chains valid under its own rules, a node follows the branch with the greatest accumulated proof of work.

So the key factor to be aware of is a chain split. When miners build a chain that is not compliant to the BIP, BIP-110 nodes can separate from the broader network. Non-BIP-110 nodes may continue following the higher-work branch, while BIP-110 nodes could remain on a compliant branch with less accumulated work. 

Corporations dealing with chain splits 

Mining companies face the highest immediate economic exposure. Electricity and machine time are sunk costs. A miner should select the branch it expects other miners, nodes, and users to recognize and mine on it. A miner may also stop mining and wait for the chain split to resolve. If BIP-110 and non-BIP-110 chains develop independently, miners must track chainwork, signaling, validity under both rule sets, and their own mining pool’s stance, and the market value assigned to each branch.

Corporations operating exchanges and institutional custody should prepare for settlement uncertainty. During an extended split, the ordinary six-confirmation standard loses much of its value because each branch can show six confirmations independently. Operators should monitor both branches, raise confirmation requirements, pause large deposits or withdrawals when risk rises, and delay final settlement until one branch has decisively accumulated more work or the transaction has sufficient depth on all viable branches. Different validation rules can produce chain splits, false confirmations, and double-spend risk.

Let’s consider a chain split occurring at block height S. 

Suppose a deposit appears on Chain A at S+4 and on Chain B at S+6. Once both chains reach S+12, the deposit has substantial depth on each branch (assuming we are still using six-confirmations). Now, this number of six confirmations should change depending on the work on each branch. And it might be the case that the number of confirmations one would like to see would be different for each branch. The main point is that the operator must wait until both branches reach the requisite confirmations. The operator can at that point be confident that the transaction remains, not matter which branch becomes canonical.

If the transaction appears on only one branch, the operator should wait for that branch to win or apply chain-specific accounting. That would be the only way to ensure no double spending happens. In practice, monetary transactions should always eventually appear on both branches, since the BIP-110 chain does not prohibit monetary transactions.

Conclusion 

The main thing to be aware of is a chain split. If there is no split, then there is nothing that needs to be done differently. Even with a chain split, BIP-110 will not create insurmountable disruptions. 

For corporations that may be impacted by a chain split, the main action to take is to lengthen confirmation times and monitor both branches. For node-running corporations that support the BIP, the main action is to start running it on their nodes, if they haven’t already. 

Miners, as usual, should direct their hashrate based on their view of which branch will end up with the most accumulated proof of work. Exchanges and custodians should lengthen settlement procedures and maintain visibility into both chains, should a chain split occur. For the daily operations of most corporate Bitcoin users, BIP-110 changes very little, if it changes anything at all. 

Disclaimer: This content was prepared on behalf of Bitcoin For Corporations for informational purposes only. It reflects the author’s own analysis and opinion and should not be relied upon as investment advice. Nothing in this article constitutes an offer, invitation, or solicitation to purchase, sell, or subscribe for any security or financial product.

This post Corporation’s Approach to the BIP-110 Soft Fork first appeared on Bitcoin Magazine and is written by Allard Peng.

Bitcoin Magazine

Quick Maths On STRC Buybacks: The Truth About Net Bitcoin Per Share and Accretion

Strategy initiated open-market repurchases of STRC last week (July 20 through July 26, 2026), buying 288,930 shares for ~$25 million at an average price of $86.52. Notably, the company bought no Bitcoin and continued to grow its cash reserve. 

So what is going on here? Why is the largest Bitcoin treasury company buying back its credit? 

Context 

In June 2026, STRC fell far below the $100 stated amount. Check out these two articles for some in depth analysis about what exactly happened: 

• Crash Post Mortem – What Happened to STRC in June 2026
• The Sixth Lever 

Last week’s STRC buyback follows Strategy’s Digital Credit Capital Framework, announced on June 29 in response to the June volatility, which authorized up to $1 billion of repurchases across STRC, STRF, STRD, and STRK. Likely because STRC is now viewed as Strategy’s flagship product, STRC was identified as the initial priority for these buybacks. 

Buyback logic starts with the position of MSTR common stock in the capital structure. Common equity owns the residual value after every senior claim has been satisfied. Strategy’s BTC and cash are its liquid assets. Debt and preferred stock sit ahead of MSTR. Strategy’s USD Reserve (read: cash) offset part of those senior claims. The common stock therefore represents the value left after subtracting debt and preferred stock from the bitcoin reserve and adding back available cash.

This is effectively Strategy’s recently introduced “Net Bitcoin Per Share” metric. Strategy’s current methodology calculates Net BTC by taking bitcoin holdings and subtracting the bitcoin-equivalent value of out-of-the-money convertible debt, other debt-like instruments, and outstanding perpetual preferred stock, then adding back the USD Reserve. Notice that this is exactly the same description as the prior paragraph! 

Net BTC is divided by fully diluted common shares to produce Net BPS. Strategy’s disclosures mark July 23 as the boundary for its revised mNAV methodology, which uses Net BPS as its denominator.

This metric gives MSTR investors a direct view of BTC economically attributable to common equity after senior claims. Gross Bitcoin Per Share can rise when Strategy issues more preferred stock or debt to buy bitcoin. Net Bitcoin Per Share captures the liability created alongside that bitcoin purchase, answering the question of how much bitcoin remains for common shareholders after the more senior investors in the capital structure are paid. 

Therefore, Net BPS provides a framework for measuring the accretive or dilutive effect of capital markets transactions on MSTR. Think of it as another new metric that investors may evaluate along with the existing metrics already being used. 

Ok, but why STRC buybacks? 

The answer is that retiring liabilities at below their notional values is accretive on a net BTC basis. 

Let’s consider a simple balance sheet with easy numbers to understand the basic mechanics.

Assume a company owns $100 million of BTC and carries $50 million of senior liabilities. Common equity is therefore a $50 million residual claim: 

$100 million assets –  $50 million liabilities = $50 million equity 

Now assume the company can retire those $50 million of liabilities for $40 million. It uses $40 million of its assets, leaving $60 million of assets and zero remaining liabilities. The common equity residual rises from $50 million to $60 million. 

$60 million assets –  $0 liabilities = $60 million equity 

The equity claim went from $50 million to $60 million. So spending $40 million to eliminate a $50 million claim creates $10 million of value for the residual owner (the common equity investor). 

The STRC repurchase follows the same structure. Strategy paid an average of $86.52 to retire a security with a $100 stated amount. Each repurchased share removed $100 from the preferred stock claim used in the company’s Net BTC calculation while consuming only $86.52 of capital. The $13.48 spread creates gross accretion to MSTR.

Strategy retired $28.893 million of STRC stated amount for about $24.998 million based on the reported average price. The difference equals approximately $3.895 million, and this value accrues to MSTR. 

(It’s worth mentioning that also related to this is STRC’s current 12% annualized dividend rate. Retiring $28.893 million of STRC stated amount also removes roughly $3.47 million of annual dividend requirements. Also consider that since STRC is still well below $100, the company likely will raise the dividend, meaning the actual annual dividend expense removed is likely higher.)

Conclusion 

Net BTC identifies the residual BTC owned by the common stock by considering all the senior liabilities which sit ahead. The STRC buyback is a move of financial engineering to improve the Net BTC per share metric of the company.

Disclaimer: This content was prepared on behalf of Bitcoin For Corporations for informational purposes only. It reflects the author’s own analysis and opinion and should not be relied upon as investment advice. Nothing in this article constitutes an offer, invitation, or solicitation to purchase, sell, or subscribe for any security or financial product. 

This post Quick Maths On STRC Buybacks: The Truth About Net Bitcoin Per Share and Accretion first appeared on Bitcoin Magazine and is written by Allard Peng.