MSTR: the STRC Backdoor

This sale comes after a silent transformation of the model. Strategy has shifted from a premium regime, where  mNAV was the central metric and where common equity issuance financed accumulation, to a financing regime, where the cost of capital is now explicit and where the equity channel has not been used in any meaningful way for several months.

Accumulation has continued through a rapid expansion of preferred issuance, starting in late 2024 and reaching nearly $15.5bn by early June 2026, largely concentrated in the STRC series. Strategy increasingly operates less as a Bitcoin proxy trading at a premium, and more as a financing structure exposed to a non-productive asset and relying on a conditional carry.

The analytical break in mNAV

From 2020 to 2024, Strategy operated an accumulation model that its supporters referred to as the flywheel. The company issued common equity above the value of the Bitcoin it held, meaning at an mNAV greater than 1. The proceeds raised were used to acquire additional Bitcoin. The balance sheet expanded. Bitcoin per share increased, reinforcing the justification for a renewed premium. The cycle then repeated.

The mechanism proved highly effective over a four year period. At its peak on 20 Nov 2024, the  mNAV based on market capitalization reached 3.29. Three key factors supported this premium, namely access to Bitcoin exposure without the need for wallet or operational custody, the leverage embedded in convertible debt, and the never-sell pledge, referring to Michael Saylor’s stated commitment not to sell the accumulated Bitcoin. 

From late 2024, the engine began to stall progressively. The  mNAV based on market capitalization declined from 3.29 times to 0.86 times as of 3 Jun, representing a 74% compression of the premium over 18 months. Three converging forces account for this dynamic. First, spot Bitcoin ETFs provided institutional investors with direct exposure to the underlying asset without the corporate risk associated with Strategy. Second, the proliferation of so-called copycat companies adopting a Bitcoin treasury model gradually eroded Strategy’s scarcity as a listed proxy, contributing to the compression of the premium. Third, Bitcoin itself has become a more institutionalized asset, structurally reducing the premium that a leveraged wrapper could command.

The  mNAV-equals-1 threshold is a simplistic reading that no longer holds under the current capital structure. For common equity issuance to be accretive to Bitcoin per share, it is not sufficient for market capitalization to exceed Bitcoin net asset value. It must exceed that value after accounting for senior claims, on a fully diluted share basis. The corresponding calculation yields, for Strategy as of 3 Jun, an estimated threshold of around 1.2 times on an enterprise mNAV basis, rather than 1.0 times on a market capitalization  mNAV basis. Strategy explicitly acknowledged this framework in its first quarter 2026 communication, presenting an Aggregate Diluted Shares Outstanding (ADSO) breakeven threshold, at which issuance becomes accretive once diluted market capitalization exceeds Bitcoin net asset value.

As of 3 Jun, Strategy’s enterprise  mNAV stands at 1.23 times, slightly above the estimated accretion threshold of 1.2 times. It has, however, fallen below this level on several occasions in recent months, which explains why Strategy has chosen not to rely on this channel in any meaningful way for the past seven months. The mNAV metric has therefore lost its function as a binary operational financing indicator. The relevant threshold has shifted toward a more stable range around 1.2 to 1.3 times, reflecting no longer simple accounting parity but the relative cost of different funding sources.

The Preferred Stack Under Close Examination

The suspension of the common equity channel has not slowed accumulation. Strategy’s Bitcoin holdings increased from 446400 BTC at the end of 2024 to 843706 BTC as of 3 Jun, representing an almost doubling over an 18-month period. This expansion has been financed through a silent shift in the funding engine, from common equity toward a stack of preferred stock, whose total liquidation preference amounts to a $15.48bn, according to the official Credit Metrics published by Strategy.

The breakdown by series as of 3 Jun reveals an extreme concentration. STRC accounts for nearly two thirds of the total liquidation preference, with a notional amount of $10.49bn. The other series combined total approximately $4bn to $5bn. Growth has been overwhelmingly concentrated in STRC since late 2025, while the other series have remained largely flat in notional terms. Over the second quarter of 2026, from late March to 3 Jun, STRC increased from $5.02bn to $10.49bn in notional, representing a rise of $5.47bn over nine weeks, equivalent to an average pace of $607mn per week. The 18 May purchase of 24869 BTC for $2.014bn was funded by more than 96% through STRC issuance in the preceding seven days, based on the observable change in notional.

Strategy’s preferred stock are different

The composition of the stack reveals a behavioral anomaly that warrants explicit identification. Strategy’s preferred instruments do not behave like institutional preferred securities, they behave more like quasi-equity instruments leveraged to Bitcoin. The correlations calculated over more than 250 trading sessions for each series illustrate this clearly. For a conventional institutional preferred, correlation with the benchmark preferred index would typically range between 0.70 and 0.85, while correlation with the underlying equity would fall around 0.30 to 0.50. Strategy’s preferred strictly invert this profile.

For STRC as of 3 Jun, correlation with MSTR equity stands at +0.70, with Bitcoin spot at +0.64, and with the S&P 500 Preferred Stock Index at only +0.29. For the three other observable series, the profile is similar. 

This has two operational implications. First, the likely buyers are not traditional preferred investors such as institutional income funds and insurance companies, but rather equity or crypto allocators seeking a leveraged carry vehicle on Bitcoin. Second, in the event of equity stress on MSTR, Strategy’s preferred will not act as the senior buffer their legal subordination would suggest, because their market pricing will already have declined in tandem with the underlying equity.

Another useful metric to assess the viability of the preferred channel is the secondary market pricing dispersion across the four series. If the market were pricing Strategy risk uniformly, the series would trade within a narrow range around par value, with differences attributable only to variations in coupon. This is not what is observed.

The analytical diagnosis is straightforward. The market no longer prices Strategy credit uniformly. It already differentiates across series based on perceived subordination levels and payment structure. STRC benefits from an anti discount structure, with a variable dividend designed to defend par, and from contractual seniority in dividend payments relative to STRK, STRD and STRE, as outlined in the STRE prospectus of November 2025. STRD, which combines the most subordinated position among the non convertible preferred with the absence of any par protection mechanism, is treated by the market as the riskiest instrument. The 14.25% yield compensates for this perception.

A rigorous reading of the model requires clarifying the exact legal nature of preferred dividend obligations. These dividends do not constitute an obligation in the same sense as bond debt. The STRC prospectus, as well as those of the other Strategy series, explicitly states that dividends are payable when, as and if declared by the board. The board may decide to defer them, partially or in full, without triggering any technical default or acceleration of any debt.

This discretionary nature of preferred payments is analytically important. The Strategy model does not represent a fixed cash obligation of $1.71bn per year, but rather a flexible cash obligation with deferral optionality. In a scenario of prolonged stress, Strategy can suspend preferred dividends without triggering default, accumulate arrears at a capped rate, and resume payments when conditions allow. However, dividend deferral would constitute a strong signal of stress, likely to materially affect the cost of capital and the market’s perception of risk.

The six layers of defence

The annual cash requirement of $1.71bn can be serviced through several channels that can be activated sequentially depending on conditions. Quantifying each layer allows for an assessment of where the true stress threshold lies.

First layer, the USD reserve. At $900mn as of 3 Jun, following the repurchase of the 2029 convertibles between 21 May and 26 May, which consumed $1.38bn, it covers 6.3 months of mandatory annual dividends. The buffer is significantly reduced compared with the $2.25bn available before the buyback, but remains material.

Second layer, the issuance of convertible debt. Strategy has historically raised convertibles at low coupons, ranging from 0 to 2.25% with a weighted average of 0.42%. This capacity has not disappeared, although the lower  mNAV makes the conversion option less attractive for investors and therefore pricing more challenging.

Third layer, common equity issuance above the accretion threshold. With an enterprise  mNAV at 1.236x, just above the estimated threshold of 1.22x, this layer is technically open but limited in capacity. A move in  mNAV above 1.3x would materially reopen this channel.

Fourth layer, additional preferred issuance. The current pace of $607mn per week on STRC demonstrates that the capacity exists, although the marginal cost is increasing.

Fifth layer, Bitcoin sales. With 843706 BTC in treasury and a current BTC price of $66 613, covering the $1.71bn of annual dividends in Bitcoin would require the sale of 25700 BTC, or approximately 3.05% of the treasury per year. 

Sixth layer, preferred dividend deferral. As outlined previously, this option is available at any time without triggering a default, with arrears accumulating at a capped rate.

Aggregate cost and the refinancing wall

The arithmetic of preferred dividends tells the full story. From zero at the end of 2024, annual preferred dividend run rate reached $797mn at the end of 2025, $1.048bn by the end of the first quarter of 2026, and $1.712bn as of 3 Jun. This represents a doubling in six months. STRC accounts for roughly three quarters of this annual cost.

The shift from common equity to preferred introduces a structural transformation. Equity is a flexible form of capital with no explicit cost. Preferred is capital with a contractual cost that remains discretionary but carries consequenThe shift from common equity to preferred introduces a structural transformation. Equity is a flexible form of capital with no explicit cost. Preferred is capital with a contractual cost that remains discretionary but carries consequences if payments are suspended, including restrictions on common dividends, governance implications, and the accumulation of arrears. The Bitcoin held on the balance sheet does not generate contractual yield but rather conditional appreciation. Strategy therefore operates in a structure where a remunerated liability finances a non remunerated asset.

The carry is not mechanically negative, it is conditional. As long as the expected return on Bitcoin over the holding horizon exceeds the marginal cost of preferred, the carry remains positive in expectation. However, as the market for Strategy’s preferred instruments becomes more segmented and the required marginal coupon rises, the risk premium absorbs an increasing share of the expected Bitcoin return.

A detailed mapping of Strategy’s convertible debt allows for the identification of the structural stress window ahead. As of 26 May 26, 2026, six convertible instruments remain outstanding for a total of $6.71bn. The effective refinancing wall is concentrated over a nine month period between September 2027 and June 2028.

Three convertibles have put dates within this window, representing a total of $4.51bn. Convert 2028 at $1.01bn with a put date of September 16, 2027, Convert 2030 B at $2.00bn with a put date of March 2, 2028, and Convert 2029 at $1.50bn with a put date of June 2, 2028. The respective conversion prices are $183, $433 and $672.

At the current share price of $137, MSTR would need to increase by between 34% and 391% depending on the instrument to neutralize the put. Absent a significant rally, cash outflows of $4.51bn should be anticipated over the September 2027 to June 2028 window, in a context where the marginal cost of preferred has increased.

The pivot toward sales, first operational signal

Three consecutive events in the final week of May 2026 materialize the narrative transition. On May 5, during the first quarter earnings call, Saylor publicly mentioned for the first time the possibility that Strategy could sell Bitcoin to fund preferred dividends. On 28 May, according to on chain data aggregated by Arkham, Strategy moved 411.6 BTC from its Coinbase custody account to a cold wallet. At the end of May 2026, Strategy confirmed the sale of 32 BTC, the first sale in the company’s history since the inception of the accumulation strategy in August 2020.

The size, amounting to 0.004% of holdings, confirms the testing nature of the move rather than a meaningful balance sheet operation. However, the precedent is now established. Future structuring of larger scale sales becomes legally and operationally feasible under the same rationale. Within the financing regime, Bitcoin can now potentially act as a source of liquidity, on par with preferred issuance or convertible refinancing. This shift in status is analytically more significant than the size of the initial transaction.

Conclusion

Strategy has transitioned from a premium trading Bitcoin proxy to a financing structure built on three distinct components, Bitcoin as the asset base, a growing stack of preferred and convertible liabilities, and common equity, which absorbs the difference and captures or bears the relative variations between asset and liabilities. This shift materially changes the analytical framework through which the vehicle should be assessed.

The model has moved from equity funded accumulation with no explicit cost to a regime where the cost of capital is now visible and rising, while  mNAV has lost its role as a central operational signal. Viability now depends both on the trajectory of Bitcoin and on continued access to funding channels.

In the near term, the structure remains resilient, supported by six funding layers, USD reserves, convertible debt, common equity issuance, additional preferred issuance, Bitcoin sales, and the option to defer dividends. The use of these different layers is not neutral and can carry meaningful implications for the company and the sustainability of the model. Market signals such as preferred pricing, issuance pace, and dividend levels are becoming critical indicators of emerging stress.

The primary risk is no longer limited to Bitcoin price or near term solvency, but increasingly relates to the evolution and sustainability of the capital structure.