Research
Why Insider Buying Is Loudest in Micro-Caps
A CEO buying $1M of her own stock means something very different at a $200M company than at a $200B one. Decades of research show insider buying predicts returns mostly in small, under-covered firms. Here's why, and where we draw the line.
When a CEO buys a million dollars of her own stock, the headline reads the same whether the company is worth $200 million or $200 billion. The signal is not the same. At the mega-cap, it is close to noise. At the micro-cap, it can be the most informative trade of the quarter.
This is not opinion. It is one of the oldest and most replicated findings in the insider-trading literature, and it is the single biggest reason ClusterDesk only watches small companies.
The Size Effect
Nejat Seyhun's foundational 1986 study of insider profits found that the abnormal returns following insider purchases are concentrated in smaller firms. Lakonishok and Lee, working with a far larger sample in 2001, reached the same conclusion: insider purchases predict future returns, and that predictive power lives almost entirely in small-cap stocks. In the largest companies, insider buying barely moves the needle.
The pattern has held up across decades and datasets. The smaller the company, the more an insider's purchase tells you.
Why Size Changes Everything
The reason is information asymmetry, and it comes down to who else is watching.
A mega-cap is covered by 30 or 40 sell-side analysts, quant funds modeling every shipment, and an army of journalists. By the time an executive files a Form 4, there is very little she knows that the market has not already priced. Her purchase is a vote, but it is a vote in an election where everyone already has the polling data.
A $300 million micro-cap is covered by one analyst, or none. No quant desk is running a model on it. Most investors have never heard of it. The gap between what the people running the company know and what the market has priced is enormous. When an insider acts on that gap, the purchase carries real information, because almost no one else has it.
That same obscurity has a second effect. In a heavily-watched stock, any genuine signal gets arbitraged away in minutes. In an under-followed micro-cap, the signal can sit there for weeks before the market notices, which is exactly the window a patient reader can act in.
The Cluster Multiplier
This is where it compounds. Alldredge and Blank (2019) found that insider purchases made within two days of a colleague's purchase generate about 2.1% of abnormal return over the following month, nearly double a solo trade.
Now layer that on top of the size effect. A single insider buying an under-covered $200 million company is already a high-information event. Several insiders independently buying that same company in the same week is the densest signal available in public markets: maximum information asymmetry, maximum internal agreement, minimum outside attention. That intersection is the entire thesis behind a cluster buy.
Where We Draw the Line
ClusterDesk only tracks clusters at companies between roughly $50 million and $500 million in market cap. Both ends of that band are deliberate.
Above $500 million, analyst coverage thickens and the insider's informational edge starts to erode. The signal is still there, but it fades into a market that is already paying attention.
Below $50 million, a different problem appears. Nano-caps are thin, easily manipulated, and often impossible to actually buy without moving the price yourself. A "signal" you cannot act on is not a signal. That range is also where promotional schemes cluster, and we would rather miss a few real ones than forward you a pump.
The band in the middle is the sweet spot: small enough that insiders hold a genuine edge, large enough that the stock is real and investable.
The Takeaway
Most insider-tracking tools treat a buy at Apple and a buy at a $250 million industrial the same way. Same table, same formatting, same alert. But the research is clear that those two trades live in different worlds. One is a footnote. The other is the closest thing public markets offer to a look at the cards.
The edge in insider data was never just which insiders buy. It is where they buy. We watch the only place the signal is loud enough to hear.
References:
- Seyhun, H. N. (1986). "Insiders' Profits, Costs of Trading, and Market Efficiency." Journal of Financial Economics, 16(2), 189–212.
- Lakonishok, J. & Lee, I. (2001). "Are Insider Trades Informative?" Review of Financial Studies, 14(1), 79–111.
- Alldredge, D. M. & Blank, B. (2019). "Do Insiders Cluster Trades With Colleagues?" Journal of Financial Research, 42(2), 331–360.