That is not quite the same as what people object to. They object to someone trading because they overhead two CEOs discussing a merger at the table next to them at a restaurant.
No amount of research can produce the same information.
That said "just overheard" is not a sure thing either, so it can be argued that it's possible to get similar quality of information from other sources. (Eg looking at their trash, looking at who goes to their office, which law firm offices do they visit.)
But this is what courts determine post hoc. If the overhearing was a baseless rumor/joke, or the merger simply went nowhere for some reason, no one will care about this proto-crime.
Also, it's up for debate for the courts which kinds of information can be determined from public sources. Is there a "holographic principle" for insider material information or not? (And even if there is the practical accessibility, computational requirements and so on make a big difference.)
> Leaving things up to the court is defensible when the ambiguities are unforeseeable or impossibly numerous, but neither apply here.
It appears you said one thing, and then the opposite.
Was that intentional? As it makes your comment seem internally inconsistent.
If there isn’t a clear line, how will legislators set demarcation points that people won’t argue in court?
However, I am suggesting that we simply not draw the line between A and B and have them both be legal because:
(1) The "bad" one is actually not that bad, both morally and efficiency-wise, especially when we clearly announce "this behavior is going to be legal, so everyone needs to be careful and protect themselves"; and
(2) there are other places to draw the line that are not so hopelessly disputable.
Since I am suggesting we put the line somewhere else where it can be made unambiguous, we are not forced to rely (nearly as much) on court interpretation.
(I'm not a lawyer, but that's what I've read.)
I'd be interested in a link to further reading if you can recommend one.
An example I saw in one article was a hedge fund that bought satellite imagery, and counted cars in parking lots to estimate how well a business was doing. That's fine, because it wasn't information that came from an insider.