This model is used as a basis for understanding the concept of 'market failure'. The market fails because food safety is under-supplied. Three specific market failures are identified. These are risk perception, information asymmetry and social costs and benefits. Risk perception is where the public’s awareness of risk is different to that of expert scientists. The public may want more safety than some scientists think is necessary. Information asymmetry is where food producers have more information about food safety risk, compared to consumers. This extra knowledge, which the industry holds, means that they can under-supply food safety without the customer knowing about it. Commercial kitchens, which are unclean and hidden from the public, provide an example of this concept. Social costs and benefits are where the activities of the food industry, affect people outside of the business. For example, where food poisoning caused by the food industry, affects the National Health Service. It is suggested that the government needs to intervene to correct these market failures. In other words, the supply of food safety needs to be increased. This is to make sure that the demand for food safety equals its supply. This leads into the main research undertaken as part of the thesis. The core issue is the extent to which government intervention is demanded by the public.
The government intervenes through the provision of public goods which, in this context, is the work undertaken through the Food Standards Agency. To estimate the demand for food safety activities, undertaken by the Agency, it is necessary to elicit the public's willingness to pay for food safety. The thesis uses a stated preference technique, to try and estimate the demand for the government’s food safety work. Stated preference, in this context, is where a member of the public is asked to state what level of additional food safety spending they would prefer. A zero option is included too. Such safety spending could be used for the employment of additional food inspectors or environmental health officers.
It is suggested that the technique is most appropriate for the valuation of common cases of food poisoning. It is less appropriate for wide-ranging and controversial topics such as Genetically Modified (GM) food. Methodological problems still arose with a restricted safety concept, covering only common cases of food poisoning. One of the main problems was ‘part-whole-bias’. This was where the relatively narrow scope of the question was overlooked. Respondents tended to generalise to include, in their valuation of the food safety issue, more serious forms of food safety hazard.