Decentralised News · Research & market intelligence
BlackRock, Goldman Sachs and Morgan Stanley provide a small, auditable case study in how an institutional-ownership headline can change when the denominator is put back.
What matters
- The three groups’ ordinary-share positions totaled 48.75 million IBIT shares at June-end, down about 1.04% from March-end.
- Their reported long-share ratio rose from 3.56% to 3.76% because IBIT’s outstanding share count fell faster.
- These are investment-manager filings, not a census of beneficial owners or a measure of unhedged Bitcoin conviction.
Scope of this investigation
Nine primary documents were read: six original manager SEC information tables, IBIT’s Q2 Form 10-Q, SEC FAQ guidance and the Form 13F instructions. Public SEC XML was downloaded for large tables that exceeded the web reader’s size limit. Rows were filtered on CUSIP 46438F101 and aggregated by manager and quarter, excluding rows with a put/call field. Extracted rows are retained with source URLs. This is a selected three-group, six-observation case study, not an all-manager census or an independently verified beneficial-ownership register. Q3 is deliberately excluded because its filing deadline had not passed at the cutoff.
Test the assumptions · Inspect the sources · Read the evidence table
The signal
A rising institutional percentage can coexist with falling reported holdings. Track units, the denominator and reporting scope together.
DN Alpha Thesis
DN extracted IBIT rows from six original SEC information tables: two quarters each for BlackRock, Goldman Sachs and Morgan Stanley. The combined ordinary-share position fell by 511,451 shares. Yet its ratio to the fund’s outstanding units rose by approximately 0.20 percentage points. The fund’s own Q2 filing explains the apparent contradiction: shares outstanding declined by roughly 6.25%. IBIT quarter-end reconciliation.
This is a three-group case study, not a completed institutional census. It uses Q2 as the latest quarter whose general filing deadline had passed at the 13 November research cutoff. Q3 filings were still arriving ahead of the 16 November deadline. The conclusion is about interpreting disclosed distribution, with a deliberately limited claim about the market as a whole. SEC filing calendar.
The share count tells a different story from the percentage
BlackRock’s ordinary IBIT share rows rose from 14,009,304 at March-end to 15,034,046 at June-end, an increase of about 7.31%. Goldman Sachs declined from 17,985,051 to 17,232,201, approximately 4.19%. Morgan Stanley declined from 17,263,778 to 16,480,435, approximately 4.54%. The directions differ within this small sample. BlackRock Q1 / Q2; Goldman Q1 / Q2; Morgan Stanley Q1 / Q2.
Those observations do not justify saying that institutions uniformly accumulated or uniformly exited. One group increased its reported position, two reduced theirs, and the aggregate moved only modestly. The result also depends on choosing these three groups; adding other managers could change both direction and concentration.
The comparison is nevertheless useful because it is reproducible. It begins with a security identifier, applies the same ordinary-share filter to each filing and sums the relevant rows within each consolidated report. It does not select a convenient dollar value from a holdings website and silently treat it as shares purchased.
Fund contraction can raise a holder ratio without a purchase
IBIT reported 1,382,480,000 shares outstanding at 31 March and 1,296,040,000 at 30 June. During Q2, 100,920,000 shares were created and 187,360,000 redeemed. The three groups’ combined position fell from 49,258,133 to 48,746,682 shares, but dividing by the corresponding fund totals produces 3.56% and 3.76%. Fund shares and creation/redemption activity.
A percentage always embeds two moving quantities. If a holder sells more slowly than the fund contracts, its ratio can rise. If a holder buys while the fund expands even faster, its ratio can fall. Neither percentage alone identifies the direction of the holder’s activity.
For a reader assessing distribution, this changes the interpretation of apparent concentration. A larger ratio could reflect new allocation, relative persistence during outflows or simply the selection of a shrinking denominator. Those mechanisms have different implications. The data here establish relative persistence for the aggregate selected position, alongside a small absolute decline; they do not establish market-wide broadening.
Options are visible, but net economic exposure is not
Goldman Sachs’ Q2 table separately reported puts on 7,910,500 underlying IBIT shares and calls on 3,276,500, alongside 17,232,201 ordinary shares. Adding all three fields would produce 28,419,201, but calling that number shares owned would mix different instruments. The ordinary-share total is the appropriate input to this article’s share-count comparison. Goldman information table.
The SEC form supplies separate put/call and discretion fields because those distinctions matter. An option’s underlying-share amount is not the number of ETF units held outright and is not a delta-adjusted net exposure. Short positions are absent from Form 13F, and reported longs are not netted against them. Form and instructions · SEC reporting guidance.
It would be equally wrong to subtract puts mechanically from shares and announce a net bearish position. The table does not provide the full set of strike prices, maturities, offsets and nonreportable exposures needed for that conclusion. A disclosed long can be part of a hedge, client portfolio, market-making activity or another strategy. Those possibilities should remain possibilities until additional evidence identifies the purpose.
A manager name does not identify the ultimate investor
The filing framework concerns investment discretion, with fields for other managers and voting authority. A consolidated group filing can include positions managed for clients or through different affiliated managers. It should not be read automatically as the named corporation placing its own unrestricted balance-sheet cash into Bitcoin. Reporting structure and instructions.
This distinction is particularly relevant when the name on a filing is also a prominent financial brand. A recognizable institution can make a position look like an endorsement by a single decision-maker. In reality, a consolidated report may combine many mandates with different objectives. The filing identifies an accountable reporting perimeter; it does not supply the investment thesis of every underlying account.
The practical safeguard is to keep three questions separate. Who reports the discretion? Which security and instrument are reported? Who ultimately bears the economic risk? The first two can often be answered from the filing. The third may require fund reports, client mandate disclosures or information that is not public. A credible article should not fill that gap with a story about institutional conviction.
What a broader distribution study would have to establish
A genuine breadth claim would need a consistent population of reporting managers across completed quarters. New and exiting reporters, acquisitions, amended filings and changes in consolidation would need to be reconciled. The researcher would also need to distinguish an existing manager adding a new position from a newly captured filing that had simply been missing from the earlier dataset.
Concentration should then be calculated within a single fund and a clearly defined reporting universe. An index calculated from three selected groups measures the balance among those groups, not concentration among all ETF holders. Combining raw share counts across different funds would add units with different economic meanings. A value-normalized analysis could answer a different question, but it would still need aligned prices and dates.
Even a comprehensive reported-long total need not be a beneficial-ownership percentage. Securities lending can support multiple gross long claims, while the SEC requires an owner that lends securities to continue reporting them and excludes short positions. A ratio above 100% is therefore a reason to investigate reporting structure and gross claims, not a number to cap automatically. Long-position and securities-lending guidance.
The benefit of a small verified case is that it exposes these problems before a large dataset makes them harder to see. The purpose is not to abandon broad analysis. It is to define the observation precisely enough that the next quarter can genuinely confirm or contradict it.
The 2027 signal is persistence through changing conditions
The constructive interpretation is that a stable core of reported holdings can remain while shorter-duration money leaves a fund. If a broad, consistently defined manager panel later shows more persistent holders, expanding ordinary-share positions and participation across distinct channels, the evidence for deeper distribution would strengthen.
The skeptical interpretation is that visible positions are concentrated in a small set of intermediaries, hedged or sensitive to arbitrage economics. If a broader panel reveals shrinking ordinary positions, fewer persistent reporters or a rising proportion of derivatives activity, the claim of widespread long-term allocation would weaken. Neither hypothesis is resolved by the three-group sample alone.
For investors, the decision is whether to treat an ownership headline as independent evidence for an allocation. This case suggests a demanding but manageable standard: require the dated units, the matching outstanding-share count and a definition of whose holdings were measured. For product distributors, the same discipline helps distinguish broader reach from a change in who remains after redemptions.
The observed result is neither a forecast of Bitcoin’s price nor a verdict on ETFs. It shows that a modest absolute decline can become a rising percentage. Any 2027 narrative about institutional adoption should survive that elementary reconciliation before it is asked to support a larger market thesis.
The evidence, side by side
| Manager / fund | 31 Mar 2026 | 30 Jun 2026 | Change | Source |
|---|---|---|---|---|
| BlackRock ordinary IBIT shares | 14,009,304 | 15,034,046 | +7.31% | S1 / S2 |
| Goldman Sachs ordinary IBIT shares | 17,985,051 | 17,232,201 | −4.19% | S3 / S4 |
| Morgan Stanley ordinary IBIT shares | 17,263,778 | 16,480,435 | −4.54% | S5 / S6 |
| Selected groups total | 49,258,133 | 48,746,682 | −1.04% | S1–S6 |
| IBIT shares outstanding | 1,382,480,000 | 1,296,040,000 | −6.25% | S7 |
| Selected reported-long ratio | 3.56% | 3.76% | +0.20 percentage points | DN calculation |
DN ETF Denominator Check
Defaults reproduce the selected three-group Q1–Q2 2026 IBIT case. User changes are scenarios. Ratios measure reported gross long shares relative to outstanding shares, not beneficial ownership; no automatic 100% cap.
The interactive calculator starts when this page loads. The worked examples remain available in the article.
All entries finite; holdings nonnegative; outstanding-share denominators positive. A zero prior position makes percentage position change not applicable.
Scenarios to watch
Observed three-group case
Default matched-quarter inputs
Position −1.04%; outstanding shares −6.25%; reported-long ratio rises from 3.56% to 3.76%.
Unchanged fund size
Use 1,382,480,000 outstanding shares in both periods
The ratio falls with the selected position. Removing denominator contraction changes the interpretation.
Flat holdings, shrinking fund
Keep 49,258,133 reported shares in both periods; retain the observed fund decline
The reported-long ratio rises without any additional shares in the selected position.
Method and limitations
Nine primary documents were read: six original manager SEC information tables, IBIT’s Q2 Form 10-Q, SEC FAQ guidance and the Form 13F instructions. Public SEC XML was downloaded for large tables that exceeded the web reader’s size limit. Rows were filtered on CUSIP 46438F101 and aggregated by manager and quarter, excluding rows with a put/call field. Extracted rows are retained with source URLs. This is a selected three-group, six-observation case study, not an all-manager census or an independently verified beneficial-ownership register. Q3 is deliberately excluded because its filing deadline had not passed at the cutoff.
The selected filings are quarter-end snapshots and do not show intraperiod trades, ultimate beneficial owners or net economic exposure. Reporting amendments and consolidation changes can require later restatement. The sample is chosen for auditable comparison, not statistical representativeness. No market-wide concentration score, institutional adoption percentage or price forecast is inferred.
Primary sources and evidence register
Source dates and reporting periods differ. Several documents from one issuer are one source family, not independent confirmations. The evidence download contains the observations used in this article.
[1] BlackRock — Form 13F Information Table, Q1 202613 May 2026 · Q1 2026 quarter-end positions
- IBIT ordinary-share rows total 14,009,304 shares.
- Put underlying-share amount 0; call underlying-share amount 227,800.
Interpretation limit: Primary SEC XML downloaded and parsed directly; extracted CUSIP 46438F101 rows saved alongside article. Options excluded from ordinary-share totals.
[2] BlackRock — Form 13F Information Table, Q2 20267 August 2026 · Q2 2026 quarter-end positions
- IBIT ordinary-share rows total 15,034,046 shares.
- Put underlying-share amount 0; call underlying-share amount 227,800.
Interpretation limit: Primary SEC XML downloaded and parsed directly; extracted CUSIP 46438F101 rows saved alongside article. Options excluded from ordinary-share totals.
[3] Goldman Sachs — Form 13F Information Table, Q1 202615 May 2026 · Q1 2026 quarter-end positions
- IBIT ordinary-share rows total 17,985,051 shares.
- Put underlying-share amount 16,281,600; call underlying-share amount 6,846,600.
Interpretation limit: Primary SEC XML downloaded and parsed directly; extracted CUSIP 46438F101 rows saved alongside article. Options excluded from ordinary-share totals.
[4] Goldman Sachs — Form 13F Information Table, Q2 202614 August 2026 · Q2 2026 quarter-end positions
- IBIT ordinary-share rows total 17,232,201 shares.
- Put underlying-share amount 7,910,500; call underlying-share amount 3,276,500.
Interpretation limit: Primary SEC XML downloaded and parsed directly; extracted CUSIP 46438F101 rows saved alongside article. Options excluded from ordinary-share totals.
[5] Morgan Stanley — Form 13F Information Table, Q1 202615 May 2026 · Q1 2026 quarter-end positions
- IBIT ordinary-share rows total 17,263,778 shares.
- Put underlying-share amount 0; call underlying-share amount 0.
Interpretation limit: Primary SEC XML downloaded and parsed directly; extracted CUSIP 46438F101 rows saved alongside article. Options excluded from ordinary-share totals.
[6] Morgan Stanley — Form 13F Information Table, Q2 202613 August 2026 · Q2 2026 quarter-end positions
- IBIT ordinary-share rows total 16,480,435 shares.
- Put underlying-share amount 0; call underlying-share amount 0.
Interpretation limit: Primary SEC XML downloaded and parsed directly; extracted CUSIP 46438F101 rows saved alongside article. Options excluded from ordinary-share totals.
[7] iShares Bitcoin Trust ETF — Form 10-Q for the Quarter Ended June 30, 20265 August 2026 · Q2 2026; quarter-end comparisons
- Shares outstanding: 1,382,480,000 at 31 March; 1,296,040,000 at 30 June.
- 100,920,000 shares created and 187,360,000 redeemed during Q2.
Interpretation limit: Use quarter-end shares in financial statements, not the later cover-page share count.
[8] SEC — Frequently Asked Questions About Form 13F6 March 2026 · Current staff guidance read 13 November 2026
- Q3 2026 filing deadline is 16 November 2026.
- Short positions are not reported or netted against long positions; securities owned but lent out remain reportable.
Interpretation limit: The date is the relevant deadline-answer update; individual FAQ entries have their own dates.
[9] SEC Form 13F — Information Required of Institutional Investment ManagersPublication date not established; see reporting period · Current form and instructions read 13 November 2026
- Separate reporting fields identify put/call, investment discretion and other managers.
- Reporting concerns investment discretion and reportable holdings; it is not a register of ultimate beneficial owners.
Interpretation limit: Instructions must be applied with filing amendments and reporting-manager relationships.
Frequently asked questions
Does a higher holder ratio mean the manager bought shares?
No. The denominator may have fallen faster than the position. Compare the absolute share count and the fund’s outstanding units at matching dates.
Why are puts and calls excluded from the share total?
They are separate instruments. Their underlying-share amounts cannot be added to ordinary shares and labeled outright ownership, or mechanically netted without more information.
Why use Q2 when some Q3 filings are available?
The Q3 2026 general filing deadline is 16 November. At this article’s 13 November cutoff, early filings would create an incomplete and potentially selected comparison.
Can the reported-long ratio exceed 100%?
It can in a gross-claims dataset. Investigate reporting relationships, lending, amendments and units; do not automatically cap it or call it a beneficial-ownership share.
