That's a big part of it. The other part is the picking. How can I, or the average person, judge the efficacy of safety protocols? I'm sure there are plenty of well-meaning people who invest enormous effort into precautions that are completely invisible to me. But I also know that the cheapest form of employment insurance is being visibly indispensable, which would tend to subsidize highly visible actions regardless of efficacy.

Then you can factor in moral hazard; people take more risks if they feel more safe. This means that if something increases safety, but less than you'd imagine, mandating it can increase danger. e.g. if drivers thought that seatblets made them 99% less likely to die, when the real number was 10%, you'd naturally expect more fatal accidents with more people driving as if their risk of death had dropped by two orders of magnitude.