Questions / Insider trading

What is insider cluster buying, and what does the evidence show?

By OQRO. Published . 4 min read. Last reviewed .

High view of the New York Stock Exchange trading floor with its steel trusses, hanging screens, trading posts, flags and a crowd of traders below, and an NYSE banner on the far wall.
The New York Stock Exchange trading floor. Photo: Carol M. Highsmith, Public domain (donated by the photographer to the Library of Congress; no known restrictions). Original

The short answer

Cluster buying is a market term for two or more insiders of the same company buying its stock on the open market within a short period; it is not a term in the SEC's rules. One published study found that clustered insider purchases were followed by abnormal returns above 2% in the next month, while OQRO's own 2026 test found that such stocks still trailed the S&P 500 in the typical case. Neither result predicts any single stock.

What the term means

"Cluster buying" is jargon for several insiders of one company, such as the chief executive and two directors, buying its shares in the open market in the same few weeks. It is not defined in the Form 4 instructions: the word does not appear in the SEC's Form 4 at all. Every site that uses it picks its own window and its own rules for what counts.

OQRO's definition is stated in its methodology: two or more insiders buying within 30 days, counting only open-market purchases of $10,000 or more, and leaving out grants, tax withholding and option exercises. A purchase is a Form 4 line with code P, as explained in Form 4 transaction codes.

Why people look at it

The reasoning is simple: one purchase can have many causes, and several people choosing to buy the same stock in the same weeks is a less ordinary event. The SEC's own investor bulletin says many investors believe reports of insiders' purchases and sales can provide useful information about their views of the company, and it adds that insiders may sell for any number of reasons, including liquidity and diversification. Sales are harder to read than purchases for that reason, which is why cluster discussions focus on buying.

What one study found

Dallin Alldredge and Brian Blank, in the Journal of Financial Research (volume 42, 2019), studied daily insider trading. Their abstract says insiders cluster trades around those of other insiders at their firm, especially insiders they work closely with, that clustering is greater when informational advantages are larger, and that clustered insider purchases were followed by abnormal returns in excess of 2% during the subsequent month. They say the results are consistent with informed trading. We read the abstract on the RePEc listing, not the full paper, and we cannot speak to its sample, its method or later studies.

What OQRO's own test found

OQRO publishes a test of its own on the methodology page, read on 9 October 2026. It took every verified open-market purchase of $10,000 or more with a 90-day result, 634 trades, and compared each stock with the S&P 500 over the 90 days after the filing.

SituationTradesBeat the S&P 500Typical result vs S&P 500
Insider purchase alone29134%minus 2.3 points
Two or more insiders buying within 30 days16443%minus 1.9 points
Insider purchase and a Congress purchase within 30 days17942%minus 5.7 points

The page's own reading is that the stocks were more often ahead of the index when several people bought, about 4 in 10 against 3 in 10, yet the typical result was still behind it, and that the gap is small and covers one short period. Read next to the study above, the two do not agree and do not need to: they measure different periods, markets, windows and definitions.

How common clusters were in one quarter

We counted distinct insiders with an open-market purchase at each company, for transactions dated from 11 July to 9 October 2026 (90 days), counted on 9 October 2026.

MeasureCount
Open-market purchase records2,417
Companies with at least one purchase456
Companies with two or more different buyers173
Companies with three or more87
Companies with five or more25
Most buyers at one company16

For comparison, 13,250 open-market sale records were dated in the same window, at 1,159 companies. This window is 90 days, longer than OQRO's 30-day cluster definition, so these are counts of companies with several buyers in a quarter, not of formal clusters.

What a cluster does not tell you

It does not say why the insiders bought, whether they discussed it, or what the stock does next. Purchases can be small, can follow a price fall, can be encouraged by company policy, or can be unrelated to each other. A cluster describes filings, and the study and the test above describe averages across many cases, not any one company.

OQRO shows these as a place to look and links every filing. The article on how to read a Form 4 shows what to check on each one.

Sources

More questions on this topic

Related guides

This explains public filings in plain words. It is not legal or investment advice. Corrections: contact page.