Having no plausible escape (or counter measures) is also a big factor for nuclear scare. If there's a fire or flood, a few weeks/months later everything will be back to normal. If radiation leaks contamination remains for generations (even if it's not that bad as news make it sound) and that freaks out people. That and the general fear of unknown and invisible (just look at the 5G panic for that)
I don't think this is a "fear" or a "feeling" thing. Accidents do have a non-negligible probability. Then it can kill people, can make land completely unhabitable. I don't have an _unsubstantiated_ "bad feeling", "fear" or "panic" about it, because accidents pose a real, even proven problem.
Furthermore, there might come a time when the knowledge about radioactivity is diminished so far that people call it a "believe" thing. And then it will cause even further problems.
Irrational part of this fear is ignoring the very low probability of it ever happening. People have this strong, but from mathematical perspective irrational preference for 100% safety, compared to say 99.99999% safety, even when the 2nd offers many benefits.
I think it is rather irresponsible to make up such a number.
With Chernobyl 1986 and Fukushima 2011, two ultimate MCA's happened in my past lifetime. Does your number include them? And, even if it is a small percentage: if it does happen, the costs involved are so much more than one would be ready to set aside (e.g., if you just measure the monetary effect, Fukushima costs $750 billion, and the US insurances would max. payout $13 billion)
I would be very happy if the world does experience the next 50 years without no more of these events. But I honestly consider that as quite unlikely. Especially from a mathematical perspective.
The number was just to illustrate the point (but I'm pretty sure the actual probability, if modern technology taken into the account, is lower than that). Also take into the account that neither of those events, no matter how much media coverage they've got, in the end didn't affect that many people outside those locations. Chernobyl did scare us in Eastern Europe, but those effects would be way smaller if it wasn't for the dysfunctional Soviet political system that first tried to hide it, so it was a very unique event unlikely to happen ever again. US had the 3 miles island accident, which was a huge thing when I was a kid, and now it's almost forgotten.
Before accounting for 40 years of technology improvements in reactor design (the 40 years with the most stunning safety improvement in humanities' history) that suggests a probability of ~0.004%.
40 years of tech improvements could feasibly have reduced the risk by 2 orders of magnitude. ivanhoe's made up number is in fact defensible. It has big error bars on it but it is not unreasonable.
