Nine of the top ten bitcoin treasury companies carry passing or Good compensation ratings. One carries Bad. The outlier is not the smallest, the newest, or the most leveraged. It is Metaplanet — holder of 43,000 BTC, the second-largest corporate bitcoin treasury on earth.
The number that explains the rating never appears in an earnings deck. It is a ratio buried in the charter: the company's option pool equals 14.7% of its fully diluted shares. The peer benchmark is 4.0%. Named executive exposure sits at 8.2%, against a peer average of 0.8% — roughly ten times the sector norm, held by the people who set the terms.
The bytecode never lies, only the intent does.
Metaplanet trades as 3350.T on the Tokyo Stock Exchange and as MTPLF over the counter in the United States. Two years ago it was a modest hospitality operator sliding toward insolvency. Then it pivoted to the digital asset treasury model Strategy popularized: issue equity, convert it into bitcoin, market “BTC per share” as the shareholder metric. The pivot worked. The equity rose roughly 1,300%, and bitcoin per share grew around 40-fold.
That is the story the company tells. The mechanism underneath tells a different one. In February 2023, as part of a rescue plan for a near-bankrupt firm, the charter was amended to create an employee option pool. The pool was not a fixed share count. It was defined as 20% of the total shares the company may issue, then wired to reset automatically.
The digital asset treasury category now spans at least ten listed companies holding bitcoin and marketing per-share accumulation. Their pitch to institutions is convergence: equity exposure with a bitcoin engine attached. Governance is the load-bearing wall of that pitch, because the engine only works if management's incentives run parallel to shareholders'. Metaplanet is the only member of that set to earn a Bad grade.
When a treasury company issues shares to buy bitcoin, the share count grows. Under the clause, the option pool grows in lockstep, because its reference is the issuable share base rather than a fixed number. Issuance triggers a proportional expansion of management's claim. No committee convenes. No shareholder vote is held. The formula runs.
The arithmetic over two years is the evidence:
- Total shares outstanding: 153.9 million → roughly 1.35 billion. A 776% expansion.
- Option pool: 46 million → 319.5 million shares. A 595% expansion.
- Option pool as a share of fully diluted equity: 14.7%, versus 4.0% for peers.
The performance gate is thin to the point of absence. The pool requires continued employment. It does not require a BTC-per-share hurdle, an earnings threshold, or vesting tied to value creation. Shareholders have never voted on the pool's growth or on revisions to its terms.
Two rollbacks followed — August 18 and September 11 — cutting the pool 41% to 188.2 million shares. Even then, roughly 105.4 million potential new shares still hang over the cap table, about 7% of the company. VanEck, which assigned the rating, listed four corrective changes. None have been implemented.
To test the mechanism rather than describe it, I modeled the pool the way I model rebasing tokens: a supply that self-adjusts against a reference. In my audit work, the recurring failure of rebasing designs is never the rebase itself but the choice of reference — whoever picks the anchor controls who gets diluted. Here the anchor is issuance. Management's claim expands with every dollar raised, whether or not that dollar bought bitcoin accretively.
Complexity is the bug; clarity is the patch.
Placed against the peer set, the ratios harden. Metaplanet holds 43,000 BTC and a 14.7% pool. Strategy, the largest treasury and the governance benchmark, carries a Good rating. BitMine and four others rated Good; three rated as passing. Only Metaplanet earns the outlier grade. Within a category competing for the same institutional allocators, that gap is structural, not cosmetic.
One figure resists clean reconciliation. A claimed 273 million shares attributed to the clause does not match the 188.2 million currently in the pool. The difference may separate issued from potential shares, or it may reflect disclosure never itemized. Either way the ambiguity is itself a signal: a mechanism this consequential should not require the reader to reverse-engineer its own accounting.
So I ran the counterfactual. Replace the automatic reset with a fixed 46-million-share pool, the 2023 starting point. The treasury strategy still dilutes — issuance continues — but management's claim stops compounding alongside it. Dilution becomes a function of the bitcoin acquisition alone, not the acquisition plus an auto-expanding insider allocation. The delta between those two worlds is the cost of the clause, and shareholders bear all of it.
The clause was not a drafting accident. It was set deliberately in a distressed moment and never revisited at scale — the pattern I see most in post-mortems. A control calibrated for a small, failing company was inherited, unexamined, by a large, successful one. Code compiles, but does it behave?
The clause lives in Japanese charter law; its economic effect reaches US over-the-counter holders and European investors under MiCA-adjacent disclosure. Regulators scoreboard disclosure, not dilution mechanics — and the mechanic is where the value moved.
The market read the two rollbacks as remediation. The data reads them as recalibration — a reduction in extraction rate, not removal of the extraction primitive. The clause still tracks issuance. The pool still resets automatically. Lowering the ceiling reduces the slope of the curve; it does not change its direction.
The deeper blind spot is decoupling. A well-designed treasury plan ties management to BTC per share — the exact metric the DAT thesis rests on. This one ties management to financing scale. In expansion the two move together and the misalignment hides. In contraction they diverge: the claim still grows with issuance while each new share buys fewer bitcoin, so BTC per share falls as insider allocations rise. The market prices hope; the auditor prices risk.
There is a subtler vulnerability. Because the clause executes automatically, accountability is diffused into a formula. A formula cannot be negligent. It cannot be subpoenaed. A board that later wants to tighten terms has no discrete decision to defend — the mechanism acted, not the people. Every edge case is a door left unlatched, and this one was built with the door on a timer.
Short-term, the rating is unlikely to move the stock; governance headlines rarely price like earnings. The durable consequence is mechanical. If index providers and ETF underwriters fold governance grades into passive inclusion criteria, Metaplanet's structural disadvantage stops being sentiment and becomes a scheduled outflow.
That asymmetry — one company in the peer set carrying the outlier grade — poses a question the cohort has avoided. Will treasury companies rewrite compensation governance before the next contraction exposes the mismatch, or does the industry need the drawdown to price a risk already legible in the numbers?
