How to read insider transactions: Form 4 codes, holdings and footnotes

Start with the transaction, not the headline

An insider disposition is a starting point for a research question. It does not, by itself, establish that an executive chose to sell shares because the business outlook deteriorated. A gift, an option exercise followed by withholding, and a sale can produce very different changes in reported holdings. The useful task is to reconstruct what happened before attaching a motive.

This guide uses a historical Berkshire Hathaway filing and a separate hypothetical worksheet. The historical example illustrates document reading, not a current recommendation about Berkshire or a statement of anyone’s present holdings. The worksheet shows how a researcher can turn several rows into a share-count bridge. Keep the reported facts, your arithmetic and your interpretation in separate columns so that another reader can reproduce the result.

Identify the person, security and relevant dates

Open the original filing in EDGAR and record the reporting person, issuer, security class and transaction date. Save its URL with your notes. A news feed’s publication date is not necessarily the date of the transaction. The signature date also should not be substituted for the filing’s acceptance timestamp when reconstructing what information was available to investors.

The SEC’s investor bulletin explains that Form 4 generally reports insider transactions, usually within two business days. That is a useful orientation, not a reason to assume every ownership change follows an identical reporting timetable. For this research exercise, read what the actual form reports rather than trying to make a legal judgment about its timeliness.

Create one row per reported transaction. If the filing includes multiple dates, preserve them individually. Combining a week’s activity into a single number can obscure an exercise followed by a sale, or make two unrelated transactions look like one decision.

Read each code in its own column

The transaction code describes the type of activity. The SEC’s bulletin identifies S as a sale, P as a purchase, M as an exercise or conversion of a derivative received from the company, and F as payment of an exercise price or tax liability using securities received from the company. G identifies a gift. Read the explanation attached to the specific row before deciding which description fits your research question.

Do not confuse that code with the acquired-or-disposed column. An entry marked D in that column describes a disposition; it is not automatically transaction code D. Likewise, D in the ownership-form column identifies direct ownership. The same letter can therefore answer different questions depending on its heading.

A practical worksheet should give these fields separate spaces: transaction code, acquired or disposed, share amount, reported price, subsequent holdings and direct or indirect ownership. Copying only the letter and amount discards the context needed to interpret them.

A historical gift shows why the distinction matters

The linked Warren Buffett Form 4 reports a May 22, 2026 transaction in Berkshire Class B common stock. Its row shows 160 shares, transaction code G, a disposition indicator D and a reported price of zero. The footnote describes a gift to a charitable organization. The row reports 1,162 Class B shares held directly afterward.

That evidence supports describing the transaction as a reported gift. It does not support describing it as a market sale for zero dollars or as evidence that the shares had no market value. The price field belongs to the reported transaction, not to a valuation of the company.

Table II separately reports Class A holdings and explains that each Class A share is convertible into 1,500 Class B shares. This is why the Class B remainder cannot be presented as the reporting person’s entire Berkshire exposure. Use the example to practice reading the form; do not carry its historical balances forward as current holdings.

Build a share bridge with a deliberately simple example

Consider a hypothetical executive who begins with 10,000 directly held common shares in one class. Assume an option exercise delivers 2,000 additional common shares, 600 of those shares are withheld for taxes, and a separate transaction sells 400 shares. Assume no other transactions, transfers or share-count adjustments. These are invented teaching inputs, not a reconstruction of any company’s filing.

The common-share bridge is 10,000 plus 2,000 minus 600 minus 400, leaving 11,000 shares. Gross dispositions total 1,000 shares, but the common-share balance increases by 1,000. Calling the executive’s entire activity a 1,000-share sale would merge tax withholding with the separate sale and hide the exercise.

Now add an assumed sale price of $50 for the 400-share transaction. Gross sale proceeds would be $20,000 before fees or taxes. Do not apply that price to the withheld shares unless their own reported valuation supports it. The worksheet does not supply an exercise price, tax rate or total compensation value, so it cannot calculate the executive’s net cash benefit.

Keep derivative changes and common shares separate

The hypothetical exercise has two related effects: an option position changes and common shares are delivered. Record both where the filing reports them, but do not count them as two independent purchases of the same economic exposure. Your common-share bridge should follow common shares; a separate derivative schedule should follow the derivative position.

Use a separate bridge for every security class and ownership category you analyze. If shares move between direct ownership and a trust, a change in the direct bucket may not be equivalent to a sale outside the person’s reported holdings. The relevant footnote and ownership description determine what you can responsibly say.

When your reconstructed ending balance disagrees with the reported balance, stop interpretation and investigate the gap. Check omitted rows, different security classes, ownership categories and the sequence of transactions. Do not insert an unexplained adjustment merely to make the spreadsheet balance. An unresolved difference belongs in the research notes as a limitation.

Use footnotes to constrain the interpretation

Read every footnote attached to the amount, price and ownership fields you use. A short table label can require a longer explanation. In the historical example, the gift footnote supplies the purpose of the reported transfer, while the conversion footnote explains why another class appears in the derivative table.

Also inspect any trading-plan indication in the original rendered form. The form’s checkbox refers to a transaction under an arrangement intended to satisfy Rule 10b5-1(c)’s affirmative-defense conditions. Its presence is not your independent finding of legal compliance, and a text extraction that prints the checkbox label may not reliably show whether it was selected. Do not guess from the label alone.

For a research memo, record only what you can see: the disclosed arrangement, relevant dates if supplied, and the transaction it covers. Leave motivation as an open question unless a reliable source addresses it. A plan reference alone cannot establish whether a stock is attractive.

Verify the document before using it as evidence

A Form 4 is a disclosure document, not a brokerage confirmation of your own purchase. The SEC’s March 2024 investor alert describes scams using fake forms to claim that investments were made or that additional payments would recover money. It advises checking filings in EDGAR and explains that the SEC does not send transaction confirmations.

For ordinary research, the same discipline is useful: start from the issuer and reporting person in the official filing system, then compare any third-party summary with the underlying document. A genuine link should still be checked for the right company, person, period and transaction. A correct document can be summarized incorrectly.

Keep the source URL beside each important observation. If a summary says “insider selling” but the source shows a gift, write down the discrepancy rather than averaging the two descriptions. The evidence should determine the label used in your memo.

Turn the filing into a bounded research conclusion

Finish with three short statements. First, state the reported transaction and its date. Second, describe the change in the specific holdings you reconciled. Third, identify what the filing leaves unresolved. In the hypothetical case, the executive sold 400 common shares in one transaction while ending with more common shares after an exercise and withholding. The assumptions do not establish a view on future company earnings.

Before changing a company thesis, connect the observation to the question you are researching. Is the concern about management incentives, a particular ownership change or a claimed signal of business weakness? Each question requires different additional evidence. Financial results and business disclosures still have to carry the analysis of operating prospects.

The repeatable output is a transparent transaction worksheet, not a bullish or bearish label. A reader should be able to trace every number back to a row, understand every assumption in the arithmetic, and see exactly where your interpretation stops.