the problem with scandal markets is how they often create further scandal and can be targeted haphazardly. the damage they do to their targets and to the platform aren't worth it when they can be wielded against innocents.

I think that the biggest case against scandal markets isn't any of your three bullet points, but rather the idea that scandal markets might drag down all of prediction markets -- scandal markets might themselves become a scandal (like how early attempts to use prediction markets at US intelligence agencies were torpedoed by the accusation of being "terrorism markets"). And in a broader sense, if the #1 application of prediction markets is "exposing the wrongdoing of powerful and influential people", then prediction markets might quickly make a lot of powerful enemies! It is in this sense that I worry that scandal markets might be net-negative. But this doesn't seem to meet your resolution criteria since you might still believe they are a "beneficial way of handling community behavior", even if that good effect comes at the expense of the wider prediction markets / forecasting ecosystem.

@JacksonWagner I'm a bit skeptical that scandal markets are effective in actually getting scandals publicized, since I don't think insider trading is particularly likely or would be noticed. My hesitance with the "powerful people will bring down scandal markets" take is that if a scandal market for a powerful person is high and they get mad, then either they'll have a scandal, in which case it will look good for scandal markets; or there would be money lying on the ground for that powerful person. I mostly believe that scandal markets are likely to be mildly net positive, though they could conceivably be quite net positive or net negative.