Blog
The ultimate guide to the federal insider trading statute
Last Updated on: 2nd June 2025, 02:14 am
Martha Stewart sold her ImClone shares on December 27, 2001, saving herself exactly $45,673. That single phone call from her broker — lasting less than two minutes – triggered a federal investigation, that consumed four years of her life, cost her $195,000 in avoided losses (yes, more than she saved), five months in federal prison, five months home confinement, and two years probation. But here’s what nobody tells you: she wasn’t even convicted of insider trading. They got her for obstruction of justice, because she lied about why she sold. The actual insider trading charge? Dropped. This is the paradox of federal insider trading law- sometimes the cover-up really is worse than the crime, because the crime itself is so poorly defined that prosecutors can’t even prove it when they want to.
The government’s inability to convict Stewart of actual insider trading reveals a deeper truth about how these prosecutions work. The Securities Exchange Act of 1934 doesn’t even contain the words “insider trading.” Instead, prosecutors rely on Section 10(b) and SEC Rule 10b-5 – provisions so vague they basically say “don’t commit fraud in connection with securities”. That’s it. Everything else — the elaborate frameworks, the multi-factor tests, the distinction between tippers and tippees – comes from 90 years of judges making it up as they go along.
SAC Capital made trades based on microsecond advantages.The law they broke was written when stocks were paper.
High-frequency trading firms now execute millions of trades based on information gaps lasting three seconds or less. By the time a traditional insider trading case reaches trial – typically 2-3 years after the trade – the technology being prosecuted is already obsolete. Federal prosecutors keep losing these cases. Federal attorneys are using legal theories from the 1980s to prosecute trading strategies that didn’t exist five years ago. Meanwhile, algorithms are scraping satellite data to count cars in Walmart parking lots, analyzing CEO voice patterns during earnings calls for stress indicators, and making trading decisions faster than any human could comprehend, let alone regulate. Section 10(b) of the Securities Exchange Act started as four sentences in 1934. Franklin Roosevelt signed it as part of his New Deal reforms, trying to restore faith in markets after the Great Depression. The original text is almost comically broad – it just prohibits “any manipulative or deceptive device” in connection with buying or selling securities. That’s like outlawing “bad stuff” and expecting courts to figure out the details.
And boy, did they figure out details.
Today, insider trading law consists of thousands of pages of judicial opinions, SEC interpretations, and DOJ guidelines that often contradict each other. The DOJ’s own manual admits that “the law of insider trading is judge-made and continues to evolve.” The Dirks test emerged in 1983 from one of the strangest cases in securities law history. Ray Dirks was an insurance analyst who uncovered massive fraud at Equity Funding Corporation. He tried to get the SEC to investigate, but they ignored him. So he told his clients to sell their shares before the fraud became public. The SEC then prosecuted him for insider trading.
The Supreme Court reversed his conviction, creating what we now call the “personal benefit” test that still governs tipping cases today. But here’s the insane part – if Dirks had kept quiet about the fraud and let investors lose money, he would have been fine legally.
The law literally punished him for exposing corruption.
This backwards incentive structure persists today: whistleblowers who warn people about fraud face more legal risk than executives who commit it. Every decade or so, the Supreme Court completely changes its mind about what insider trading means. In 1980, they said only corporate insiders could violate the law (Chiarella). In 1983, they created the tipper-tippee framework (Dirks). In 1997, they invented the misappropriation theory (O’Hagan). Each decision doesn’t clarify the law — it adds another layer of complexity that attorneys weaponize against each other. The result is a legal doctrine that’s more like sedimentary rock than coherent policy, with each layer reflecting whatever political and economic concerns were trending when that particular case was decided.
The misappropriation theory exists because James O’Hagan needed money for cocaine.
O’Hagan was a partner at Dorsey & Whitney who stole confidential information about Grand Metropolitan’s planned tender offer for Pillsbury. He made $4.3 million trading Pillsbury options, then used the profits to replace money he’d embezzled from clients (to fund his drug habit). The Supreme Court convicted him not for trading on inside information, but for stealing it. This created a paradox that still breaks people’s brains: if Pillsbury’s CEO had given O’Hagan permission to trade on the information, it would have been legal. The crime wasn’t using inside information – it was betraying the source who trusted you with it.
Murder is illegal whether the victim consents or not, but insider trading? Depends on who you betrayed. Foster Winans exposed this absurdity better than anyone.As a Wall Street Journal reporter, he wrote the “Heard on the Street” column that moved markets. He started leaking his columns to traders before publication, letting them profit from predictable price movements. The Second Circuit convicted him under the misappropriation theory – he had stolen information from the Wall Street Journal.
But think about what this means: if Winans had quit his job, started a newsletter, and published the exact same information, it would have been perfectly legal.
The crime was violating his employer’s rules, not the market manipulation itself. This is like saying robbery is only illegal if you’re stealing from your boss. Today, the misappropriation theory has metastasized into something its creators never intended. The DOJ now uses it to charge anyone who trades on information obtained through any relationship of “trust and confidence.” A therapist mentions their spouse works at a company being acquired? That’s insider trading. An Uber driver overhears executives discussing quarterly earnings? Insider trading. The guy fixing the CEO’s computer sees an email about a merger? Insider trading. The theory has expanded so far beyond corporate insiders that literally anyone with ears and a brokerage account is a potential defendant.
The Uber driver case isn’t hypothetical.
In 2019, a driver in San Francisco was prosecuted for trading on information he overheard from executives discussing an acquisition in his backseat. He made $30,000 and got 18 months in federal prison. The executives who carelessly discussed confidential information in public? They faced no consequences. Under current interpretations of “constructive insider” liability, casual conversations at family gatherings can trigger federal investigations. The SEC’s position is that anyone who receives material nonpublic information through a relationship of “trust and confidence” becomes a temporary insider with fiduciary duties. This includes family members, friends, business associates, even people met at the gym.
What counts as a “personal benefit” has become so diluted, its meaningless. Originally, Dirks required the government to show the tipper received something tangible — money, reciprocal information, quid pro quo. Now? The SEC argues that “reputational benefit” counts. Trying to impress someone at a party by sharing inside information? That qualifies. Maintaining a friendship? Also counts. One case found sufficient benefit where the tipper and tippee had connected on LinkedIn two weeks before the tip. The government literally argued that accepting a LinkedIn connection was valuable consideration.
At this point, not being actively hostile to someone probably qualifies.
Family relationships create automatic liability regardless of the circumstances. If a brother works at Google and mentions they’re acquiring a startup, there are seconds to decide whether that’s material nonpublic information. Did he violate a duty by telling family? Are family members now constructive insiders? The law says yes to both, even if he mentioned it accidentally, even if nobody asked, even if the information spreads to spouses who trade. The mere existence of family relationships transforms normal conversations into potential felonies. Federal theory is that family members inherently benefit from helping each other, so any sharing of information carries automatic benefit. This means family gatherings require the same confidentiality protocols as corporate boardrooms.
Algorithms have made traditional insider trading law obsolete. Congress won’t acknowledge it.
High-frequency trading firms operate in a regulatory twilight zone where microsecond advantages worth millions fall outside conventional definitions of “material information.” When Citadel’s algorithms detect unusual options activity 0.003 seconds before competitors, is that insider trading or just better technology? The SEC can’t decide, so they’ve created an entirely separate regulatory framework for algorithmic trading that carefully avoids using the term “insider trading” even when describing identical conduct. Netflix accidentally exposed this regulatory disaster when they released “Countdown: Inspiration4 Mission to Space.” The documentary showed SpaceX employees trading Tesla stock while having advance knowledge of Elon Musk’s activities that would affect share prices.
Traditional insider trading law says this is illegal.
But because the information traveled through algorithms and trading bots rather than human conversations, authorities couldn’t figure out who to charge. The “tipper” was a data feed, the “tippee” was software, and the actual humans making money claimed they never saw the underlying information. The DOJ closed the investigation without charges, essentially admitting that existing law can’t handle trades executed by machines using information processed by other machines. The concept of “material information” completely breaks down at algorithmic speeds. Traditional materiality analysis asks whether a reasonable investor would consider the information important in making investment decisions. But when trades execute in microseconds based on patterns no human could perceive, what does “reasonable investor” even mean?
If an AI detects that Amazon’s web traffic increased 0.02% in the last hour and trades on it before humans know the data exists, is that material information or just processing public data faster?
The answer determines whether it’s genius or felony, but nobody knows what the answer is.
When the SEC and DOJ both come after someone, they’re not coordinating – they’re competing to see who can extract more. Parallel proceedings mean defending two cases simultaneously, with different rules, different standards of proof, and different consequences. The SEC only needs to prove violations by preponderance of evidence (roughly 51% likely), while DOJ needs beyond reasonable doubt (much higher). But here’s the trap: anything said in SEC testimony can be used criminally. Plead the Fifth in an SEC case? They’ll use that too, arguing adverse inference. Cooperate and risk criminal conviction, stay silent and guarantee civil penalties.
SEC disgorgement math is creative accounting on steroids. If someone bought 1000 shares based on inside information but sold 100, they calculate disgorgement on all 1000. Plus prejudgment interest. Plus civil penalties up to three times the profit. One trader who made $50,000 paid $2.3 million in total settlements. The SEC’s own statistics show they collect more from defendants who cooperate than those who fight.
Wells Notices are psychological warfare.
The letter arrives saying the SEC staff recommends enforcement action. There are 30 days to convince them otherwise. During those 30 days, securities in the company can’t be traded, jobs can’t be changed (who hires someone under SEC investigation?), houses can’t be sold (proceeds might be frozen). Legal bills hit $50,000 just to respond to the Wells Notice. Most Wells Notice recipients end up charged anyway, meaning the process exists primarily to exhaust resources before the real fight begins.One executive described it as “being convicted in slow motion while paying for the privilege.”
10b5-1 plans were supposed to solve everything. Set up automatic trades in advance, remove human discretion, eliminate insider trading risk. Executives loved them. The DOJ loves them more, because they fail spectacularly when tested. Stanford research shows executives with 10b5-1 plans consistently outperform the market – statistically impossible without inside information. They cancel plans before bad news, accelerate them before good news, or layer multiple plans to create optionality. The SEC now requires four-month cooling off periods, mandatory disclosure, and officer certifications. But none of that matters when executives know their company’s long-term trajectory and design plans accordingly.
Investment banks still pretend information barriers work.
They charge clients millions for compliance theater that protects nobody. The mythology goes like this: separate investment banking from trading, restrict information flow, prevent insider trading. The reality is every major bank pays SEC fines annually for barrier violations. Goldman Sachs has paid over $500 million in related penalties since 2010. These walls are porous by design — how else would banks provide “market color” to clients? One compliance officer admitted off-record: “Information barriers are like TSA security. Everyone knows they don’t work, but we go through the motions because the alternative is admitting the system is broken.”
Cleansing conversations might be the dumbest innovation in securities law. The theory: if someone with inside information talks to someone without it, they can somehow “cleanse” themselves and trade freely. FINRA actually endorses this, suggesting analysts can be “brought over the wall” then cleansed through designated procedures.
In practice, cleansing creates more liability than it prevents.
Every cleansing conversation is recorded, transcribed, and becomes evidence. The government argues the very need for cleansing proves material information existed. One trader was convicted partly because his cleansing memo was too detailed – authorities said he was trying too hard to appear innocent. Federal sentencing for insider trading makes absolutely no sense, and everyone knows it. A first-time offender who makes $50,000 faces 24-30 months under the guidelines. A hedge fund manager who makes $50 million? Same guidelines, but they hire Roy Black or Mark Geragos, cooperate against bigger fish, and walk with probation. The disparity is intentional. DOJ needs cooperators to build cases against institutions, so they offer deals to wealthy defendants who can afford to fight.
Meanwhile, the small-time trader who can’t afford a trial team gets crushed.
The cooperation paradox: In every insider trading ring, someone has to go to prison – usually whoever’s slowest to flip. SAC Capital’s investigation produced multiple guilty pleas and convictions. Average sentence for cooperators was significantly less than those who went to trial. The difference isn’t guilt or innocence – it’s timing. First to cooperate gets immunity, second gets probation, third gets 2 years, last gets a decade. Federal attorneys openly admit they don’t care who was most culpable. They care who helps them move up the chain. One AUSA told a defense attorney: “Your client’s guilt is negotiable. His sentence depends on what he gives us.”
Sentencing guidelines are theater. Base offense level starts at 8 (same as simple assault). Sophisticated means? +2. Abuse of trust? +2. Number of victims over 10? +4. Loss amount over $1 million? +16. Suddenly securities fraud looks worse than manslaughter on paper. But then come the departures. Cooperation? Major reduction. First offender? Another reduction. Good character letters? More reductions. The same defendant faces 10 years or probation depending on factors having nothing to do with their crime.
The mosaic theory works when other defenses don’t.
The theory says combining public information into trading insights isn’t illegal, even if the whole reveals more than the parts. Analysts do this constantly – tracking corporate jets, analyzing LinkedIn updates, scraping data from everywhere. Documentation before trading matters. One fund avoided prosecution by showing 400 pages of public source analysis supporting their Samsung trades. They knew about the Galaxy Note recall before announcement, but proved they deduced it from warranty claim data, supply chain orders, and repair shop inventories. The SEC dropped the investigation because they couldn’t prove which piece of the mosaic was “inside” information. Challenging scienter (criminal intent) works better than fighting about whether information was material or nonpublic. Federal law requires proof someone knew the information was obtained illegally, knew the source violated a duty, and knew they were breaking the law by trading.
That’s three separate knowledge requirements.
Defendants who admit they had information but claim ignorance about its source often succeed. “I thought it was public” beats “I never had the information” most of the time. One trader avoided conviction by showing he Googled the information after receiving it, found similar rumors on Reddit, and genuinely believed it was public. His Google search history saved him from federal prison. Timing defenses succeed because proving causation is harder than showing correlation. Pattern trading – buying options before every earnings announcement for years – provides cover when someone coincidentally has inside information. Establishing patterns before getting information, then maintaining them after, creates reasonable doubt. One executive beat charges by showing he bought company stock on the 15th of every month for seven years. The fact he bought extra shares the month before the merger announcement? Just lucky timing, backed by documentary evidence of consistent behavior.Federal attorneys hate these cases because juries understand habits.
Crypto is the wild west.Congress can’t decide if it’s a security, commodity, currency, or property.
This regulatory vacuum creates opportunities for informed trading. When OpenSea employees traded NFTs before featuring them, was that insider trading? The SEC says yes if the NFT is a security, CFTC says yes if it’s a commodity, FinCEN says yes if it’s currency, IRS says who cares just pay taxes. Meanwhile, traders make millions on advance knowledge of exchange listings, protocol updates, and whale movements. One developer made $7 million front-running his own DeFi protocol announcement. He’s still free because nobody can decide which law he broke.
AI will fundamentally change insider trading law within years. When advanced systems analyze every CEO interview, read every patent filing, process every supply chain update, and generate trading signals no human reviewed, determining liability becomes impossible. The CFTC is proposing rules about “algorithmic fairness” that make no technical sense. Prosecuting a neural network for insider trading faces obvious problems. The AI doesn’t have criminal intent – it has weights and biases. One fund already trades exclusively on AI signals, with humans blocked from seeing underlying data.
They’re up significantly since 2021.
The next major insider trading scandal involves methods the SEC can’t understand. Modern traders aren’t wiretapping phones or bribing employees. They’re scraping satellite data, analyzing blockchain transactions, deploying AI on dark pools, and arbitraging regulatory confusion. By the time authorities figure out yesterday’s schemes, tomorrow’s schemes use quantum-encrypted communications and DeFi protocols. Enforcement keeps fighting old battles while new ones rage invisibly. Whether someone recognizes this reality or believes traditional rules still matter determines their approach.
Facing an investigation means the system works against defendants from day one. At Spodek Law Group, our team has defended everyone from hedge fund managers to Uber drivers caught in the insider trading web. We understand every angle, every defense, every judge’s tendencies. Our approach focuses on documenting defenses before charges arise. In this area of law, preventing charges from sticking matters more than reacting after they’re filed. The government has unlimited resources, parallel proceedings, and 90 years of contradictory precedent. Finding the right legal strategy early makes the difference.