How to Read an ETF Holdings File

Key Takeaways

Direct answer: An ETF holdings file is the fund’s own daily list of every position that will feed its next net asset value calculation. SEC Rule 6c-11 requires each ETF to publish that list prominently on a free, publicly available website before regular trading opens on its primary listing exchange, and to include five fields for every holding: ticker symbol, CUSIP or other identifier, description of the holding, quantity held, and percentage weight in the portfolio. Read the as-of date first, because the file reflects the prior business day’s close, then read the non-equity lines, because cash, futures, currency, and collateral often carry weights that change how the equity lines should be interpreted.

• The daily holdings file is a rule-mandated disclosure, not a courtesy. Rule 6c-11(c)(1)(i) sets out the five required fields, and Rule 6c-11(c)(2) requires that the holdings shown are the fund’s portfolio as of the close of business on the prior business day.
• Five fields is the floor, not the ceiling. Most issuers add market value, shares outstanding, sector, country, coupon, and maturity. None of those extras is standardized across providers, so a column name means whatever the issuer’s own documentation says it means.
• The same webpage that carries the holdings must also carry the fund’s net asset value per share, market price, and premium or discount as of the prior business day’s end, plus a premium and discount history table, a line graph, and a median bid-ask spread.
• Percentage weights frequently sum to something other than 100. Rounding, negative cash, unsettled trades, and derivative notional conventions all push the total off, and each has a different diagnosis.
• The daily file is not the same disclosure as Form N-PORT. Under the currently effective rule, N-PORT reports for each month in a fiscal quarter are filed no later than 60 days after that quarter ends, and only the third month’s report is made public on filing.
• A holdings file tells you what a fund owns. It does not tell you how easily the fund could sell it. The liquidity classification a fund assigns each investment under Rule 22e-4 is reported to the SEC on Form N-PORT and is expressly not made public in fund-identifiable form.
• Two funds tracking the same index can publish materially different files on the same morning because of sampling, fair-value adjustments on foreign holdings, and different cash and futures conventions.

What Is an ETF Holdings File?

An ETF holdings file is a machine-readable or downloadable list of every security, asset, and other position an exchange-traded fund holds, published by the fund’s own sponsor on its own website. It is the most granular routine disclosure any pooled investment vehicle makes, and it is the reason ETF research can be done from the outside rather than inferred from performance.

The disclosure exists because of a specific SEC rule. Rule 6c-11 under the Investment Company Act gives exchange-traded funds the exemptions they need to operate at all: it lets an ETF share be treated as a redeemable security, lets dealers transact at market-determined prices rather than at net asset value, and lets certain affiliated parties deposit and receive baskets. Those exemptions come with conditions, and the first condition on the list is daily portfolio transparency.

The rule’s language is worth reading closely, because it constrains three separate things: what must be disclosed, where, and when. Each business day, an ETF must disclose prominently on its website, which must be publicly available and free of charge, and before the opening of regular trading on the primary listing exchange, the portfolio holdings that will form the basis of the next calculation of current net asset value per share.

Three practical consequences follow. First, the file cannot sit behind a login or a paywall. Second, it has to be up before the opening bell, not at some point during the day. Third, it describes a specific portfolio: the one the next NAV will be struck against. That last point is the one most readers skip, and it is where most misreadings start.

Which Five Fields Does the Rule Require?

Rule 6c-11(c)(1)(i) enumerates the required fields for each portfolio holding. The list is short, and everything beyond it is voluntary.

Notice what is missing. The rule does not require market value in dollars, sector, country, credit rating, coupon, maturity, or a currency code. Most large issuers publish several of those anyway, and they are useful, but they are provider conventions rather than regulated fields. If your research process depends on a column that the rule does not require, that process will break the first time you add a fund from a different sponsor.

The identifier field deserves particular care. It is the only field designed to be joined on, and it is also the field most likely to be inconsistent across sponsors. If you are comparing two funds by hand, matching on ticker plus description is usually fine. If you are building anything automated, resolve to one identifier standard first and treat name matching as a fallback that must be reviewed, never as the primary key. The same discipline underpins the weighted-overlap math in the ETF overlap analyzer, which explains why two funds can look 40 percent identical on names and 60 percent identical on weights.

Why Does the As-Of Date Matter So Much?

The single most common misreading of a holdings file is treating it as a live portfolio. It is not. Rule 6c-11(c)(2) is explicit: the portfolio holdings that form the basis for the ETF’s next calculation of current net asset value per share must be the ETF’s portfolio holdings as of the close of business on the prior business day.

So a file you download at 9:00 a.m. on a Wednesday describes Tuesday’s closing portfolio. If the fund traded during Tuesday’s session, those trades are in it. If the fund trades during Wednesday’s session, they are not, and you will not see them until Thursday morning. Over a long weekend or a market holiday, the lag stretches accordingly.

That lag matters differently depending on what the fund does:

• A broad, low-turnover index fund. One day of staleness is almost irrelevant. The portfolio on Tuesday and the portfolio on Wednesday are the same portfolio with slightly different prices.
• An index fund inside a reconstitution window. One day of staleness is the entire story. During an index rebalance the fund may be trading a meaningful share of its book, and consecutive daily files are the only public record of that transition.
• An actively managed ETF. One day of staleness is the manager’s edge, and the reason some active strategies took years to arrive in ETF form. The file tells you where the manager was, not where the manager is going.
• A leveraged or inverse fund. The file describes a portfolio that is reset daily by design, so a single day’s snapshot says almost nothing about the exposure a holder experienced over a week. See leveraged and inverse ETF risks for why the compounding, not the holdings list, is the dominant term.

A practical habit fixes most of this: before reading a single line of the file, find and record the as-of date the issuer stamped on it, and check it against the date you expect. Files served from a cache, mirrored by a data vendor, or saved into a spreadsheet weeks ago all look identical to a fresh download except for that one field.

The Lines That Are Not Stocks: Cash, Futures, and Collateral

A file of 500 equity lines and 6 non-equity lines will still have its interpretation decided by the 6. These are the entries that most often confuse a first reading.

• Cash and cash equivalents. Almost every fund carries some. A small positive balance is ordinary operational float from dividends received, creation activity, or a pending trade. A large balance is a signal worth explaining before you assume anything else about the fund.
• Negative cash. A negative line is not necessarily leverage. It frequently reflects unsettled purchases, where the securities are already on the books and the payable is not. It can also reflect a genuine borrowing. The description text is what separates the two, and if the description does not say, the fund’s annual report will.
• Futures contracts. The quantity column holds contracts, not shares, and the weight column may reflect market value, notional exposure, or the value of posted margin depending on the issuer’s convention. An equity index future is frequently used to keep incoming cash invested between the moment it arrives and the moment it is deployed, which is why a fund can hold a futures line without being a futures strategy.
• Currency forwards. A currency-hedged fund will show forward contracts alongside its securities, often with both a long and a short leg per currency. The forward legs are the hedge, and they are the reason the fund’s return differs from the unhedged version of the same index. The mechanics are covered in international ETFs, currency risk and hedging.
• Repurchase agreements and collateral. Common in short-duration and cash-management funds. The counterparty and the collateral type belong in the description, and if they are absent, the fund’s filings are the place to look.
• Securities on loan. Some issuers flag lent positions in a separate column, others do not. The presence of a securities-lending programme is disclosed in the fund’s registration documents whether or not the daily file mentions it.

The general rule: read the non-security lines first, decide what portion of the fund is not in the market the way you assumed, and only then start reading the equity or bond lines. A fund that is 4 percent cash and 3 percent futures is not a 93 percent equity fund. It is very likely a 100 percent equity-exposure fund that is holding 3 percent of that exposure synthetically, and the two readings lead to completely different conclusions about tracking and about risk.

Where the Daily File Sits Among a Fund’s Other Disclosures

The daily website file is one of four routine portfolio disclosures. They differ in timeliness, in depth, and in whether they are audited, and confusing them produces bad research.

Form N-PORT is where the daily file is most often misunderstood, so it is worth being precise about the current requirement. Rule 30b1-9 requires a monthly report of portfolio holdings on Form N-PORT, current as of the last business day or last calendar day of the month, and provides that reports for each month in a fiscal quarter must be filed with the Commission no later than 60 days after the end of that fiscal quarter. Form N-PORT’s own General Instruction F then handles public availability: information reported for the third month of each fund’s fiscal quarter is made publicly available upon filing, and the SEC does not intend to make public the first and second months’ information in a form identifiable to a particular fund or adviser.

Two research consequences follow. Public N-PORT data is a quarter-end snapshot delivered with a lag of up to two months, which makes it excellent for structure and useless for timeliness. And several fields are never public at fund level at all, including the fund’s highly liquid investment minimum, its derivatives exposure figures, its median value-at-risk numbers, and the liquidity classification it assigns to each portfolio investment.

Public N-PORT filings are searchable through SEC EDGAR full-text search, which is also where the annual report lives when you need an audited schedule of investments rather than a same-week snapshot.

Worked Example: Reading a Holdings File Line by Line

The file below is a hypothetical illustration built for this guide. It is not a real fund, and the numbers are chosen to show the specific reading problems described above. The fund is imagined as a 300 million dollar US large-cap equity ETF, and only the lines that need explaining are shown.

Four readings come straight out of those six lines.

1. Beta Systems is a 6.21 percent position, not a 5.12 percent position. The two share classes are separate lines with separate identifiers, and a concentration check that reads the file by ticker will understate the fund’s exposure to that single company by more than a full percentage point. Multi-class issuers are common in large-cap indices, and this is the most frequent single error in do-it-yourself concentration analysis.
2. The fund is fully invested even though only 96.07 percent of the weight is in stocks. Add the futures line at 1.31 percent to the equity weight and the market exposure lands at roughly 97.4 percent, with the cash sitting behind the futures as margin and operational float. Reading the cash line as "the manager is defensive" would be wrong here. The futures line is the tell.
3. The negative line is not leverage. Net other assets and liabilities at minus 0.38 percent is the reconciling entry that turns the sum of position values into net assets. Unsettled purchases, accrued expenses, and dividends receivable all land there.
4. The weights sum to 96.07 plus 1.31 plus 1.31 minus 0.38, which is 98.31 percent, so 1.69 percent of the fund is in the lines this extract omitted. That is the correct conclusion. An extract is not a file, and any weight arithmetic run on a partial download is arithmetic on an unknown denominator.

Every figure above was computed for this illustration and can be reproduced from the table. None of it is a quoted market price, a projection, or a recommendation.

Why Do the Percentage Weights Rarely Sum to 100?

Once you start adding up weight columns, you will find they almost never total exactly 100.00 percent. There are five ordinary explanations and one that should worry you.

The last row is the one worth building a habit around. Diffing consecutive daily files on the quantity column, not the weight column, is the cheapest portfolio-activity signal available to an outside researcher, because quantity changes only when the fund trades or when creations and redemptions move the whole book proportionally. Weight changes every day whether the fund trades or not, because prices move.

What a Holdings File Cannot Tell You

Daily transparency is unusually good disclosure by the standards of pooled vehicles, and it still leaves real questions unanswered. Knowing the boundary keeps you from over-reading the file.

• It does not tell you how liquid the portfolio is. Under Rule 22e-4, a fund classifies each portfolio investment as a highly liquid, moderately liquid, less liquid, or illiquid investment, using thresholds measured in days to convert to cash without significantly moving the price. Those classifications go to the SEC on Form N-PORT and are not made public at fund level. You can estimate liquidity from the holdings, but you cannot read the fund’s own answer.
• It does not tell you what the fund paid. There is no cost basis, so realized and unrealized gains are invisible from the file. The annual report carries that.
• It does not tell you why. An index fund’s holdings are an output of a published methodology; an active fund’s are an output of a process you cannot see. Neither is explained in the file.
• It does not tell you the fund’s costs. Expense ratio, spread, and the drag from internal trading all live elsewhere. See expense ratios and the total cost of ETF ownership, and the fee drag calculator for the compounding effect.
• It does not tell you how well the fund tracks. Holdings can match an index perfectly and the fund can still lag it, which is exactly the distinction drawn in tracking error and tracking difference.
• It does not tell you whether the fund will still exist next year. Fund closure risk is an asset-level and sponsor-level question, handled as a discrete step in the ETF due diligence framework.

One more limit is easy to miss. The rule requires the file to be published, not to be published in any particular format. A CSV, an XLSX, an HTML table, and a PDF all satisfy it. If your process assumes a stable schema from a given issuer, that assumption is a convention you are relying on, not a right the rule gives you.

A Repeatable Workflow for Reading a New Fund’s File

The steps below are ordered so that each one can invalidate the ones after it, which saves work.

1. Find the as-of date and confirm it is the prior business day. If it is older, you have a cached copy or a stale mirror.
2. Count the rows and compare to the fund’s stated holdings count. A mismatch means you have an extract, and everything downstream is wrong.
3. Sort by weight descending and read the top 10. This is the fastest read on what the fund actually is, and it is often at odds with the fund’s name.
4. Filter for negative weights and for blank tickers. That single filter surfaces cash, payables, futures, forwards, and anything else that is not a listed security.
5. Group by issuer, not by line. Collapse multiple share classes and multiple bond series onto the parent issuer before you draw any concentration conclusion.
6. Sum the weight column and diagnose the gap using the table above rather than assuming rounding.
7. Check the same page for NAV, market price, premium or discount, and median bid-ask spread. Rule 6c-11 requires all of those on the same publicly available site, so there is no excuse for reading holdings without them.
8. Pull the same file on three consecutive days and diff quantities if turnover, reconstitution behaviour, or creation activity matters to your question.
9. Only then compare funds. Two files that have each passed steps 1 through 7 can be compared. Two raw downloads cannot.

If the comparison step is where you are headed, the ETF overlap analyzer already implements the weighted overlap and look-through arithmetic, including the issuer-grouping problem that step 5 describes. It is the right destination for "how much do these two funds actually duplicate," and this guide is the right starting point for "can I trust the two files I am about to feed it."

Common Mistakes When Reading Holdings Data

• Treating the file as real time. It is a prior-close snapshot by rule, not by accident.
• Reading only the top 10. The top 10 tells you the fund’s character. The tail tells you its concentration, its liquidity profile, and its index methodology.
• Counting share classes as separate companies. The single most common source of understated single-name exposure.
• Ignoring blank-ticker rows. Those rows are frequently the most informative lines in the file.
• Assuming the weight denominator. Net assets and total assets diverge whenever a fund carries payables, and the rule does not specify which one the issuer used.
• Matching two funds by security name. Free-text descriptions differ across sponsors for the same security. Use identifiers.
• Inferring liquidity from position size. A 0.4 percent weight in a thinly traded small cap can be a larger liquidity problem than a 5 percent weight in a mega cap.
• Comparing a daily file to an N-PORT filing and calling the difference "turnover." Different dates, different scope, different purpose.

Reading the File Well: What Actually Changes

Daily portfolio disclosure is the structural feature that makes exchange-traded funds researchable in a way that most pooled vehicles are not. A mutual fund shareholder generally sees a full portfolio twice a year in the shareholder reports and a quarter-end snapshot through public N-PORT data. An ETF holder can see the portfolio every single morning, for free, before the market opens, because Rule 6c-11 makes that disclosure a condition of the exemptions the fund depends on to exist.

Using that advantage well comes down to three disciplines, and none of them is technical. The first is reading the timestamp before the data, because a file’s as-of date determines what question it can answer. A prior-close snapshot answers "what did this fund own," and it never answers "what is this fund doing right now." The second is reading the file’s exceptions before its bulk: the blank tickers, the negative weights, the derivative lines, and the reconciling entries decide how the ordinary equity lines should be interpreted, and they take thirty seconds to find. The third is grouping by issuer before drawing any conclusion about concentration, because a file is a list of instruments and risk attaches to companies.

Those three habits also mark the boundary of what this document can do. The file will not tell you what the fund charges, how closely it tracks its benchmark, how easily its positions could be sold under stress, or whether the sponsor intends to keep it open. Each of those is a separate question with a separate disclosure behind it, and each is covered by a separate guide in this section. What the holdings file gives you is the raw, unmediated, daily fact of what the fund owns, published on a fixed schedule in a defined minimum format. Read the timestamp, read the exceptions, group by issuer, and it is one of the most reliable primary sources available to a retail investor anywhere in public markets.

If your next question is how much of that portfolio sits in a handful of names, continue to ETF concentration risk, which takes the file you have just learned to read and turns it into a measurement.

Frequently Asked Questions

References

This guide is based on the operative SEC rules and forms, each retrieved and verified on 22 August 2026:

• eCFR: 17 CFR 270.6c-11, Exchange-traded funds: the daily website disclosure condition, the five required per-holding fields in paragraph (c)(1)(i), the prior-business-day requirement in paragraph (c)(2), the NAV, market price, premium and discount, and median bid-ask spread items, and the definitions of basket, creation unit, authorized participant, and custom basket used throughout.
• eCFR: 17 CFR 270.30b1-9, Monthly report: the requirement to file monthly portfolio holdings on Form N-PORT current as of the last business day or last calendar day of the month, and the 60-day post-fiscal-quarter filing deadline described above.
• SEC: Form N-PORT: General Instruction F on public availability, which makes the third month of each fiscal quarter public upon filing and withholds fund-identifiable information for the first and second months, together with the highly liquid investment minimum, derivatives exposure, value-at-risk, and liquidity classification items.
• eCFR: 17 CFR 270.22e-4, Liquidity risk management programs: the four liquidity classification categories and their day-count definitions referenced in the limits section.
• SEC: EDGAR Full-Text Search: the filing archive where public Form N-PORT reports and audited annual reports are retrieved.

The holdings-file extract in this guide is an original, hypothetical illustration constructed to demonstrate multi-class issuers, synthetic exposure, and reconciling entries. It is not a real fund, a quoted price, a projection, or a recommendation. The SEC adopted amendments to Form N-PORT on 28 August 2024 that would change the reporting frequency and public-availability schedule described above; as of this guide’s publication date the Commission has delayed those amendments’ effective date to 17 November 2027, with compliance dates of 17 November 2027 for fund groups with net assets of one billion dollars or more and 18 May 2028 for smaller fund groups, so the currently effective regime is the one described here. This is educational content, not personalized investment, tax, or legal advice.