Unusual 'put option' trading on airline and insurance stocks before September 11 proves that insiders had foreknowledge of the attacks

The case, at its strongest

The insider-trading claim deserves to be stated at full strength, because unlike some 9/11 claims it starts from a real anomaly rather than a misreading. The core fact is genuinely arresting: in the narrow window before the attacks, someone loaded up on bets that these exact airlines would fall, and then these exact airlines fell. Put volume on UAL and AMR was not merely up, it was multiples of normal, concentrated in the very days a person with foreknowledge would have chosen.

They bet the airlines would crash, days before the airlines crashed. Stated that plainly, the suspicion almost makes itself.

The claim can also point to something more than intuition. A peer-reviewed study in The Journal of Business examined the pre-attack option market against decades of historical data and found the put activity in the airlines statistically unusual. Proponents read that as scholarly confirmation. And there is the matter of the missing paperwork: years later the SEC acknowledged that the working files behind the 9/11 Commission's key footnote had been destroyed. To a skeptic already inclined to distrust the official account, a striking anomaly, an academic paper calling it “unusual,” and vanished evidence add up to a case that at least demands an answer.

It does demand an answer. The rest of this file is that answer, because each of those three pillars means something narrower than it first appears.

What the trades actually were

The reason the insider-trading claim collapses is not that regulators waved it away, but that they did the opposite: they ran down individual trades to the buyer, and the buyers turned out to be ordinary. The Securities and Exchange Commission opened a sweeping review, joined by the FBI, the options exchanges, and foreign regulators, and the 9/11 Commission later summarized the result in its report.

Take the single most cited number, the UAL puts of September 6. The Commission found that about 95 percent of them were bought by one US-based institutional investor with no conceivable connection to al-Qaeda, as part of a larger trading strategy. The decisive detail is what else that strategy contained: the same investor was buying calls and shares, positions that gain when airlines rise and lose when they fall. A trader who knew hijacked jets were about to destroy these companies would never hedge the bet that way. The mixed book is the signature of a routine directional or hedging play, and it is flatly inconsistent with foreknowledge.

The AMR spike of September 10 had an equally mundane root. Much of it traced to a US options-trading newsletter that had recommended exactly those put trades to its subscribers days before, a fax that went out to a mailing list, not a signal passed from a terror cell. When a tip sheet tells thousands of readers to buy a put, a volume spike in that put is what you would expect to see, attacks or no attacks.

The largest block of “suspicious” puts came bundled with calls and shares, the one combination a person with foreknowledge would never buy.

The other names dissolved the same way. Morgan Stanley, the insurers, the reinsurers: on inspection their apparently well-timed positions were pre-existing hedges, trades opened before any plausible tip window, or volume that was elevated but explicable. After all of it, the SEC's conclusion was blunt, that it had not developed any evidence that anyone with advance knowledge of the attacks traded on it. No account, foreign or domestic, was ever linked to the plot.

Why 'unusual volume' is weak evidence

The deeper lesson is statistical, and it is why the academic paper proves less than it seems. Allen Poteshman's 2006 study did find the pre-9/11 put activity in the airlines high relative to a historical benchmark. But “statistically unusual” only means the volume sat out in the tail of the normal distribution. Tails are not empty in ordinary times; unusual-but-innocent volume happens constantly, driven by earnings jitters, sector news, a widely-read newsletter, or a big fund rebalancing a hedge. Poteshman himself drew no conclusion of informed trading and noted the pattern was consistent with normal market behavior.

Two well-known reasoning traps do the rest of the work for the claim. The first is survivorship, or the Texas sharpshooter: after a catastrophe, investigators comb thousands of securities for whatever traded oddly beforehand, and in a market that large some names will always look prescient by chance. Draw the target around the bullet holes after the fact and you can find a “pattern” in pure noise. The second is base rates: even genuinely unusual volume is far more often caused by mundane things than by insider crime, simply because mundane causes are so much more common. A flag that fires on both is not proof of the rare cause.

This is why regulators do not stop at “the volume was high.” They ask who traded, and what else that person held, and when the position was opened. Those questions turn an eye-catching aggregate number into a set of individual stories, and in this case every story was ordinary. Unusual volume raised the question. It could never, by itself, answer it.

The destroyed files, and how the myth survives

If the trades were explained by 2004, why does the claim persist into the present? Part of the answer is the destroyed documents. When a researcher later asked the SEC for the working files behind the Commission's footnote, the agency said they had been purged under a routine retention schedule. For proponents this landed as vindication: the evidence was gone, so the innocent explanation must have been a cover. But routine destruction of closed-case work-papers is ordinary bureaucratic practice, and it cuts both ways: the files that would have confirmed the mundane explanations are gone too. Missing paperwork leaves a real transparency gap, which honest coverage should admit. It is not a finding of foreknowledge.

The rest of the myth's durability is structural. The claim is short, it supplies a villain and a motive, and it hides inside the technical fog of options markets that most readers cannot audit. It also never travels alone; it rides with the broader inside-job narrative and borrows its emotional charge. Later papers by conspiracy-aligned authors have claimed to detect “informed trades” in the same data, but they reargue the tail of a distribution without ever producing the one thing that would matter: a named account, held by someone tied to the plot, that traded on the secret. Twenty years of searching has not produced it.

Where the evidence lands is therefore clear, and it can respect both halves of the truth. The elevated put volume was real, and calling it “unusual” is fair. The inference that it proves foreknowledge is not fair: it was investigated, traced trade by trade, and found at every turn to have an innocent cause, with the single largest block of puts sitting inside a book of calls and shares that no plotter would ever have built. On the record as it stands, the insider-trading thesis is debunked.