Insider Trading Statistics (2026)

Insider trading is two very different things sharing one name. The legal kind, corporate officers and directors buying and selling their own company's shares and disclosing it, happens constantly and is closely tracked. The illegal kind, trading on material non-public information, is what makes headlines (see the table below).

The legal disclosures generate a genuine, if modest, investing signal: decades of research find that insider buying predicts small outperformance. The illegal side is rarer to prosecute than it is to occur, and enforcement has actually thinned in recent years even as detection technology has improved.

The line is about information, not about who you are. Company insiders can legally trade their own stock as long as they are not acting on material non-public information and they report it: purchases and sales are filed with the SEC on Form 4, typically within two business days, which is how research and services track them.

Illegal insider trading means trading (or tipping others to trade) on confidential, market-moving information, whether you are an executive, a lawyer, an IT contractor, or a friend of a friend. The maximum criminal penalty is 20 years in prison and a $5 million fine for individuals (see the penalties table below).

Through 2025, corporate insiders leaned heavily toward selling. The CEO buy/sell ratio sat around 0.36 late in the year, meaning far more selling than buying, and one service reported cumulative insider sales running about 79% above their 10-year average (these are secondary aggregator figures).

Mega-cap technology accounted for an outsized share: billionaire founders and executives reportedly took more than $16 billion off the table in 2025. Jeff Bezos alone sold about 25 million Amazon shares for roughly $5.7 billion, and Mark Zuckerberg sold around $733 million of Meta in the first quarter.

Yes, but modestly and asymmetrically. In Lakonishok and Lee's classic 1975-1995 study, stocks that insiders bought in small-cap firms earned about 7.4% abnormal returns over the following 12 months, while stocks insiders sold showed no meaningful underperformance (see the chart and table below).

Jeng, Metrick and Zeckhauser reached a similar verdict: insider purchase portfolios beat the market by more than six percentage points a year, but insider sales carried no reliable signal. Note these are older cohorts, historical results, not a promise about the future.

The asymmetry has a simple logic. There is really only one reason an insider buys more of their own stock with their own money: they think it is going up. That makes purchases informative about the insider's private read on the business.

Selling is noisier. Insiders sell for dozens of reasons that have nothing to do with the outlook, diversification, a house purchase, taxes, tuition, or simply cashing in vested equity. Because most selling is driven by liquidity rather than information, sales predict future returns far more weakly than buys do.

Not all insider buys are equal. Cohen, Malloy and Pomorski showed that more than half of insider trades are 'routine', made at the same time every year, and carry no predictive power at all. The signal lives entirely in 'opportunistic' trades that break an insider's usual pattern.

A strategy that traded only on opportunistic insiders earned value-weighted abnormal returns of about 82 basis points a month, roughly 10% a year (see the chart above). Tellingly, those same opportunistic traders pulled back when the SEC publicized new insider trading cases, a sign they knew the difference between a hunch and a tip.

The prosecuted cases are the tip of the iceberg. A UTS study by Talis Putnins and Vinay Patel, covering 21 years of data, estimated that illegal insider trading occurs ahead of roughly 1 in 5 M&A deals and about 1 in 20 quarterly earnings announcements (see the table below).

Crucially, they put the probability that any given instance is detected and prosecuted at only about 15%, implying at least four times more insider trading happens than regulators catch. Prevalence around takeovers runs about four times higher than around earnings, because merger news moves prices the most.

Enforcement has been trending down, not up. The SEC brought just 34 insider trading actions in fiscal 2024, about 8% of its standalone cases, and 32 in fiscal 2023, versus 58 back in 2012 and a modern low of 32 in 2019 (see the chart and table below).

That sits well below the roughly 50-cases-a-year historical pace. Across all categories, FY2024 produced 431 standalone actions and a record $8.2 billion in financial remedies, but insider trading has been a shrinking slice of a shrinking caseload.

Fiscal 2025 marked a sharp retreat. The SEC filed 313 standalone actions, its fewest in a decade and down 27% from FY2024, and collected just $808 million in total monetary remedies, the lowest since 2012 and down 45% year over year (see the table above).

Insider trading remained a stated priority, with cases like a March 2025 international ring charged with more than $17.5 million in illegal profits, but the overall drop reflects a change in enforcement posture rather than a change in how much trading is happening.

The agency increasingly leans on data. The Enforcement Division's Market Abuse Unit runs an Analysis and Detection Center, launched in 2011, that mines trading and market data to flag suspicious patterns, then works backward to the people behind them.

In July 2022 alone, that unit generated cases against nine individuals across three schemes tied to more than $6.8 million in gains. Tips help too: the SEC received roughly 24,000 whistleblower tips in FY2024 and paid more than $255 million in awards, though most tips concern fraud broadly, not just insider trading.

The downside is severe. Criminal convictions carry up to 20 years in prison and fines up to $5 million for individuals (and up to $25 million for corporations), a range extended by the Sarbanes-Oxley Act after earlier statutes in 1984 and 1988 (see the table below).

On the civil side the SEC can seek disgorgement of 100% of the illegal profit plus a penalty of up to three times that profit, so total civil exposure can reach roughly four times what the trade made, on top of officer-and-director bars and injunctions.

A handful of cases defined the modern era. Galleon founder Raj Rajaratnam was convicted in 2011 on 14 counts, sentenced to 11 years, and ordered to pay more than $150 million; the wiretap-driven investigation produced 80-plus related convictions (see the table below).

SAC Capital pleaded guilty in 2013 and paid a record $1.8 billion, with eight employees convicted, including Mathew Martoma over a $276 million trade on confidential Alzheimer's drug-trial data. Further back, Ivan Boesky's 1986 case (a $100 million fine, three years in prison) inspired the Gordon Gekko archetype.

One flavor of insider-adjacent trading is fully legal but deeply unpopular: members of Congress trading individual stocks. Per Unusual Whales, in 2024 Democratic lawmakers' portfolios rose about 31% on average and Republicans about 26%, both beating the S&P 500's 24.9% (though only about half of the roughly 100 active traders beat the index).

The public wants it stopped. A 2023 University of Maryland survey found 86% of voters favor banning individual-stock trading by members of Congress, including 87% of Republicans and 88% of Democrats (see the chart below). Support held for applying the same rule to the President, Vice President, and Supreme Court.

Insider filings are a legitimate, free data point, not a magic signal. The research is consistent that cluster buying by insiders, especially opportunistic, out-of-pattern purchases in smaller companies, tilts the odds slightly in your favor, while insider selling tells you almost nothing. Treat a big insider buy as one input among many, not a green light.

And resist the fantasy of the illegal edge: it is rarer to get away with than it looks, the penalties run to decades in prison and multiples of any profit, and the durable way to build wealth is a diversified, thesis-driven portfolio you rebalance over years, not a tip you act on in an afternoon.