MSCI's Final Rule on Non-Operating Companies: How the Two-Filing-Period Test Slows Strategy's Possible Removal

On 15 October 2026 MSCI published its final methodology for what it calls non-operating companies, replacing the interim restriction it had applied to firms whose digital assets make up at least 50% of total assets. Headlines said the rule could remove Strategy, the largest corporate holder of bitcoin, from the world's most widely tracked equity benchmarks. That is possible. But the detail that matters most is a persistence test: a company already in an MSCI index is removed only if it fails the screen in each of its two most recent annual filing periods. The rule is designed to be slow, and the timeline below shows where the real decision points are.

What MSCI decided

According to Benzinga, companies are first screened for substantial operating assets. Those that fail face five financial-ratio tests and are excluded if four or more are flagged. The interim measures for companies with digital assets of at least 50% of total assets end, and those companies move under the new methodology. The framework covers investment funds, business development companies and passive vehicles, which are already excluded, plus corporate issuers structured similarly.

The Block, reporting the same day, adds the rules for movement in and out. Non-constituents are ineligible for addition if they fail the screen in either of their two most recent annual filing periods. Current constituents are removed only if they fail in both. MSCI may use the most recent interim filing where one exists. Buffers apply to both the ratio thresholds and the timing of inclusion and removal, which MSCI says are meant to reduce index turnover.

On timing, The Block says the rules take effect with the November index review and that MSCI will publish a final list of potentially affected companies on 11 November. Benzinga says the framework takes effect after the close of 30 November. The two are consistent: the list on 11 November names the companies, and the changes take effect at the end of the month.

What is final and what is not

Neither of the 15 October articles we read lists the five ratios or their cutoffs. The detail we have is from MSCI's August consultation document, which was a proposal. It set a first screen under which a company passes automatically if operating assets exceed 50% of total assets. A company that fails that screen is tested against five flags:

• Operating asset intensity (operating assets divided by total assets) below 20%.
• Expense intensity (operating expenses divided by total assets) below 5%.
• Operating cash flow below zero.
• Non-operating fair-value changes above 5% of total assets.
• Capital dependence (financing cash flow divided by total assets over the latest twelve months) above 20%.

Four or more flags make a company ineligible. For existing constituents the consultation proposed looser deletion thresholds, 10% for operating asset intensity and 30% for capital dependence. We have not found confirmation that the final thresholds match the proposal, so treat every number in the list above as the August draft, not the final rule.

Why the two-filing-period rule slows everything down

MSCI's consultation explained the design choice directly: it favours a slow screen that requires persistent evidence, so that a one-off threshold miss does not trigger removal, and it argued that a faster, reactive screen would raise index volatility and turnover. The result is an asymmetry. A company outside the index is blocked after one failed filing period under the final rule, while a company inside it needs two. It is easier to keep a seat than to win one.

The consultation also described a watchlist for constituents that fail only the latest filing, with removal following if they fail again the next year. The 15 October reports do not describe the watchlist mechanics in the final rule, so we do not assume they carried over unchanged, but the structure, a warning period before a deletion, is the logic of the two-period test.

MSCI also illustrated the proposal on ACWI IMI data as of May 2026. It produced three deletions, Strategy, Yellow Cake and Metaplanet, and three watchlist entries, SharpLink, Center Laboratories and Lydia Holding. Because deletion under the proposal required failure in two consecutive annual periods, our inference is that on MSCI's own illustrative data Strategy was a deletion and not merely a watchlist name. That inference does not carry over automatically to 11 November. The list will depend on which filings count at review time and on the final thresholds. And the direction of travel has not been linear: Crypto Briefing reported in August that MSCI had backed away in February 2026 from excluding digital-asset treasury companies.

Why a bitcoin treasury company fits the profile

Strategy holds 848,250 BTC, valued by Benzinga at $69.50 billion, with unrealised gains of nearly $6 billion. Those figures imply roughly $81,900 per bitcoin at the time of writing, consistent with the $82,068 Benzinga quoted for bitcoin. They also imply a cost basis of about $63.5 billion, or around $74,900 per coin on average. That second pair is our arithmetic from rounded figures, not a reported number.

The screen does not care about size. It asks what a balance sheet is made of and how the company funds itself. Operating asset intensity captures how much of the balance sheet is operating assets. Fair-value intensity captures how much reported results come from revaluing holdings. Capital dependence captures how much of the balance sheet is being built from outside financing. A company whose assets are overwhelmingly bitcoin, whose results swing with bitcoin's price, and which finances more bitcoin purchases by issuing securities is the profile these ratios were built to find. That is a statement about how the screen is constructed. We did not obtain Strategy's actual ratios, so we cannot say how many flags it would trigger.

What exclusion would do, and what nobody has sourced

If a company leaves an MSCI index, passive funds tracking that index must sell it, and some active managers benchmarked to it may follow. The amount depends on the stock's weight and on how much money tracks each affected index. We did not find a sourced calculation of the selling pressure from the final rule. August headlines carried estimates in the range of a couple of billion dollars, but we could not trace them to a primary calculation, so we do not repeat them as fact.

Market reaction on the day was modest. Benzinga reported that Strategy's shares closed 5.62% higher at $151.29 on Tuesday and were down 2.91% in pre-market trading after the announcement. Bitcoin was around $82,000, up about 1% over 24 hours. Neither figure tells us much about the long-run effect of the rule.

The dispute

Strategy has criticised the proposal and urged MSCI to withdraw it. It calls the non-operating screen a pretext for excluding bitcoin treasury companies, and argues that MSCI's operating and non-operating classifications have no basis in accounting rules or any recognised legal framework (Benzinga). The Block reports that Executive Chairman Michael Saylor and CEO Phong Le have described MSCI's approach as discriminatory against digital-asset treasuries. Le has also said that removal would have an immaterial effect on the company, per Benzinga, a position that sits uneasily with the intensity of the objections.

MSCI's stated rationale, from the consultation, is that non-operating companies create value by holding non-operating assets, generate little cash from business activity, depend on market movements and external capital, and rely less on their own operations for growth. On that view the screen is neutral: it applies the same ratios to funds, business development companies and corporate holding companies.

The evidence offers some support to each side. The May illustration was not limited to bitcoin: Yellow Cake is UK-listed, and SharpLink is described as an Ethereum treasury firm. A rule written in ratios is also more transparent than a discretionary judgement. But the thresholds are set by a private index provider, and the classification does not come from accounting standards, which is Strategy's core point. A fair test of neutrality is whether the final thresholds catch non-crypto companies with similar balance-sheet profiles, and the 11 November list will show some of that.

Dates to watch

• 11 November 2026: MSCI publishes the final list of potentially affected companies (The Block).
• After the close of 30 November 2026: the framework takes effect (Benzinga).
• Companies' next annual filings: under a two-period test, each new filing resets whether a company has failed once or twice.

For investors the useful reading is that the rule creates a process, not a verdict. Whether Strategy is deleted, placed on a watchlist or left alone depends on filings and thresholds that become visible in the next four weeks.