Abstract
In a previous paper, we discussed that the application of cost benefit analysis (CBA) often incurs setting a value to a statistical human life (VOSL). This led to decades of research into what a reasonable value should be. These evaluations of the VOSL lead to widely varying results. Rather than attempting to harmonize on an average with large margins of uncertainty, the conclusion can be drawn that indeed there is no law of nature that determines what risk is acceptable and that, therefore, a consistent valuation of a human life cannot be expected. Nor can it be expected that there is a universally valid number for the acceptability of a risk. We argue that one should accept that standardization of acceptable risks has its practical limitations given by the – lack of – similarity in nature of the activity and the nature of the risk. In fact, attempts to force standardization are counterproductive. In many cases, one has to accept the only available alternative not involving violence, which is a political debate, terminated by the more general rule of law or constitution on how to settle such a debate and then accept the decision.