I am not convinced by arguments that people are inherently aware of risks that experts have not been until some recent number-crunching on expanded and cleaned historic data: rare disasters fare badly, I think.
Some theories along the lines of prospect theory sound much more plausible to me, I just don't accept the notion that people are aware that the daily swings accrue more pain if they are loss averse than the joy of the gains, even if the latter dominate the long run. Still, these are nice models.
I wanted to have a model where the key is the eternal delay of entering, because people want to make it at the right time, and it leads to inaction. Alas, I am not good enough with the tools about inaction to have it properly set up, and especially seeing what error of judgment is needed, if any, and what kind would be acceptable (but plausible).
