The Evolution of Risk Perceptions Related to Bovine Spongiform Encephalopathy—Canadian Consumer and Producer Behavior

Abstract

In this study the dynamics of risk perceptions related to bovine spongiform encephalopathy (BSE) held by Canadian consumers and cow-calf producers were evaluated. Since the first domestic case of BSE in 2003, Canadian consumers and cow-calf producers have needed to make decisions on whether or not their purchasing/production behavior should change. Such changes in their behavior may relate to their levels of risk perceptions about BSE, risk perceptions that may be evolving over time and be affected by BSE media information available. An econometric analysis of the behavior of consumers and cow-calf producers might identify the impacts of evolving BSE risk perceptions. Risk perceptions related to BSE are evaluated through observed market behavior, an approach that differs from traditional stated preference approaches to eliciting risk perceptions at a particular point in time. BSE risk perceptions may be specified following a Social Amplification of Risk Framework (SARF) derived from sociology, psychology, and economics. Based on the SARF, various quality and quantity indices related to BSE media information are used as explanatory variables in risk perception equations. Risk perceptions are approximated using a predictive difference approach as defined by Liu et al. (1998). Results showed that Canadian consumer and cow-calf producer risk perceptions related to BSE have been amplified or attenuated by both quantity and quality of BSE media information. Government policies on risk communications need to address the different roles of BSE information in Canadian consumers' and cow-calf producers' behavior.

Acknowledgments

We thank the Alberta Prion Research Institute for providing funding support for this project.

Notes

¹The person who contracted vCJD was actually exposed to the disease in England.

²This previous experience may come from learning the result of foreign BSE cases or other types of emergent risks which have happened.

³Media attitude indices should also be constructed and incorporated. However, it is subjective to judge the tone of BSE-related articles. Further, there is a high correlation between the negative-attitude media index and the gross media index about BSE. Given these problems, the media attitude indices are not incorporated in the analysis.

⁴It is assumed that there are no other factors leading to structural changes. In this study, the structural changes due to factors other than BSE were tested and not supported by the data.

⁵The other possible meaning of BSE include Bombay Stock Exchange, Budapest Stock Exchange, Boston Stock Exchange, breast self-examination, Bulgarian Stock Exchange, Bulgarian software enterprise, Building Services Engineering, Baku Stock Exchange, Bahrain Stock Exchange, Biological Systems Engineering, breast self-exam, etc.

⁶The word “government” may also represent governments of other countries. However, every article has been read thoroughly to screen out other foreign governments in terms of BSE impacts. The same screening is done for other words such as “scientist,” “farmer,” etc.

⁷In most of the BSE-related articles, the words “farmer” and “risk” (or “headache,” “concern,” “threat,” “fear,” “loss,” “devastate,” “disaster”) appear together in one sentence or connected sentences to express the “devastated” status of farmers due to BSE. Every article is read carefully to screen out other meanings of the two words not concerned with farmers' suffering from BSE. “Farmer” is used loosely to represent producers and ranchers, among others, although the use of the word “farmers” dominated media coverage.

⁸The BSE media coverage has low correlations with pork expenditure share (ρ = −0.02) and turkey expenditure share (ρ = −0.009). Therefore, BSE information might not have had a big impact over pork and turkey expenditures.

⁹According to the “2008 Statistics of the Canadian Dairy Industry—Complete Publication” from Canadian Dairy Information Centre, the culling rate of dairy cows in 2008 is 23%. The dairy cows culled in Eastern Canada in 2008 are therefore computed by: dairy cows culled in Eastern Canada in 2008 = dairy cow inventory in Eastern Canada in 2008 × 23%. The percentage of slaughter cow supplies from dairy industry in Eastern Canada in 2008 is estimated by: dairy cows culled in Eastern Canada in 2008/slaughter cow supply in Eastern Canada in 2008.

¹⁰Zero homogeneity has been imposed here.

¹¹In Jarvis's model, the optimal age is slaughter age, while in cow-calf operations in Canada, producers usually sell their animals to backgrounders, feedlots for fattening or packing plants for slaughtering.