Risk involves consideration both of the consequences of accidents and the frequency with which the accidents occur. Indeed formally risk is equal to the product of frequency and consequences. The important question of the perception of risk by the public and by the professional is first addressed. Two tenets are proposed as being a suitable summary of the public requirement:
1. If it can happen, then it must not matter.
2. If it matters, then it must not happen.
A mathematical interpretation is placed upon these tenets and is shown to be consistent with various professional safety targets. The tenets do not indicate what numerical values for risk would be acceptable to the public but they do show how the consequences of accidents should diminish as the frequency or likelihood of a particular accident increases. It is argued that the best way of determining what level of risk the public accepts is to be guided by statistics for man-caused accidents. These, it transpires, pose risks which are considerably greater than those implied, for example, by the professional targets for nuclear reactors. The risk posed to the public by two energy installations is summarised. The one installation, situated on Canvey Island, exports energy in the form of gas, some of which (methane) is pumped into a national gas grid. The other installation, the Sizewell “B” Pressurised Water Reactor nuclear power station has not yet been constructed, but a comprehensive risk assessment has been undertaken, the results of which are summarised. The two installations are comparable in the sense that each exports a power of the order of a million kilowatts (in the form of gas in the one case and electricity in the other). Both have been the subject of Public Inquiries. The risks posed by the Canvey installations are accepted, since they only constitute a small fraction of the risks which the public run in any case during the course of their everyday lives. The predicted risks for the PWR are smaller still. The form taken by the risks posed by both installations corresponds broadly with the two tenets. That is to say the greater the consequences the lower should be the frequency of a particular accident.