Megan McArdle suggests that insider trading may be a “victimless” crime:
Of course, if our insider buys shares, someone else doesn’t get them … but there’s no way of knowing whether that person would have held on to the stock until the merger was announced [at which point the stock is likely to be more expensive – JC]. So the insider could possibly profit without making anyone noticeably worse off…
This is wrong. If Insider X buys a stock (or anything else) at a price of $20 on January 10, and sells it at $50 on January 20, then we know that, were it not for Insider X’s participation in the market, someone else (or multiple other people) would have held the stock during those 10 days, thus achieving the $30 profit. Mcardle seems to fixate on whether “that person” (i.e. a particular person) would have held it for 10 days. That is irrelevant. A stock cannot be ownerless. A crime with an unidentifiable* victim is different from a victimless crime.
[update 11/15 – I now believe the below note is unnecessary, confusing, and maybe wrong]
*Note that the victim may be “unidentifiable” in the strong sense that there is actually no specific victim or victims, not the weak sense that we simply don’t have strong enough investigative powers to find the victim. The victim is that person who “would have” owned the stock if it weren’t for the insider’s purchase. Thus the victim is a victim only because in the hypothetical world where the insider did not act, the victim would have been better off. It is impossible to determine what exactly would have happened were it not for the insider’s action; what we do know, however, with reasonable certainty, is that someone other than the insider would have been better off in this hypothetical world than in the actual one in which the insider acted.