← Back to Trades · Updated April 2026 · 10 min read

Does Insider Buying Predict Stock Performance? What the Research Says

The idea is intuitive: if the CEO of a company spends millions of her own dollars buying stock, she probably knows something the rest of the market doesn’t. But does the data actually support this intuition? The short answer is yes — and the body of evidence is substantial.

Six Decades of Academic Evidence

Insider trading research is one of the most studied anomalies in financial economics. The foundational work dates back to Jaffe (1974) and Seyhun (1986), and the findings have been remarkably consistent across decades, geographies, and market conditions.

Jaffe (1974) was among the first to demonstrate that insider purchases generate abnormal positive returns over the following months, while insider sales provide a weaker but still meaningful negative signal. This asymmetry — buys being more predictive than sells — has been replicated in virtually every subsequent study.

Seyhun (1986, 1998) analyzed tens of thousands of SEC filings and found that stocks purchased by insiders outperformed the market by approximately 4-8% annually on a risk-adjusted basis. His work also established that the signal is strongest for open-market purchases by top executives, and weakest for sales and derivative transactions.

Lakonishok and Lee (2001) confirmed these findings with a larger dataset and found that the predictive power of insider buying is concentrated in smaller companies where information asymmetry is greatest. They estimated that a portfolio mimicking insider purchases in small-cap stocks generated 7.4% annual alpha.

Why Insider Buying Works as a Signal

The predictive power of insider buying stems from a fundamental information asymmetry. Corporate insiders possess knowledge about their companies that outside investors simply cannot access:

When an insider commits significant personal capital to buying stock, they are implicitly saying: “Given everything I know about this business, the current stock price undervalues the company.”

Key finding:The predictive power of insider buying increases with (1) the dollar amount of the purchase, (2) the seniority of the insider, (3) the number of insiders buying simultaneously (cluster buys), and (4) the stock’s recent underperformance. These four factors form the core of InsiderBrief’s InsiderScore™ methodology.

The Numbers: What Returns Can You Expect?

Across the academic literature, the consensus findings for portfolios that mimic insider purchases are:

These are risk-adjusted returns, meaning they account for market beta, size, and value factors. The insider buying signal generates genuine alpha that cannot be explained by standard factor models.

When Insider Buying Is Most Predictive

Not all insider purchases carry equal weight. Research identifies several conditions that amplify the signal:

Limitations and Caveats

No signal is infallible, and insider buying is no exception:

How InsiderBrief Uses This Research

The academic literature provides a clear framework: insider purchases predict positive returns, especially when they are large, clustered, contrarian, and made by senior executives. InsiderBrief’s InsiderScore™ methodology translates these academic findings into a practical scoring system. Every Form 4 filing is evaluated against these evidence-based factors, and only the highest-scoring trades are surfaced in our daily intelligence briefs.

The result: you get the benefit of decades of academic research, applied systematically, delivered to your inbox every morning — without needing to parse a single SEC filing yourself.

Disclaimer: InsiderBrief provides informational content and analytical tools for educational purposes. Nothing on this site constitutes investment advice, a recommendation to buy or sell any security, or an offer to transact. Always conduct your own due diligence and consult a licensed financial advisor before making investment decisions. Past insider trading patterns do not guarantee future stock performance.
/to search