Files

Abstract

Prion diseases have raised concerns in consumer’s minds about food safety associated with meat world-wide. Bovine Spongiform Encephalopathy (BSE) and Chronic Wasting Disease (CWD) both exist in Canada and consumer markets for beef, bison, elk and deer may have been affected by the diseases. While numerous studies have examined Canadian consumer beef purchasing behavior in the presence of BSE (Lomeli (2005), John(2007)), no examination of the impact of the animal diseases on consumer behavior for households who consume bison, elk and venison as part of their protein intake has been undertaken. In this study, meat consumption behavior for these specific households is examined, in particular, examining meat substitution possibilities between the exotic meats and domestic meats such as beef, pork, chicken, turkey and seafood. CWD was discovered on a Canadian elk farm in 1996 (CFIA) and since has been found on deer and elk farms and in the wild in Saskatchewan and Alberta. While to date no bison have been found with BSE, many of the trade bans put in place at the time of BSE have affected the bison market and consumers may associate bison with BSE through media coverage of the disease and trade barriers. In addition to examining the possible links between food safety concerns and consumption, the demographic characteristics of the households who consume exotic meats, and their responses to relative prices and other economic variables are modelled. No other empirical study exists on Canadian revealed meat purchasing behavior focused on these specific meats. Hobbs et al (2006) examined the potential market for bison in Canada. There are a number of past studies investigating the impact of food safety information reported in the media and product recalls on food demand. For many recent studies of consumers’ behavior associated with meat demand and food safety concerns, the linear approximate almost-ideal demand system (AIDS) model has been popular and used. In this study, the behaviour of selected households that include bison, elk and venison meat as part of their total meat consumption will be examined. The revealed impact of BSE and CWD incidences on these household’s behavior will be examined using a linear approximate almost-ideal demand system (LA/AIDS) model. Using theoretical concepts, model development can be explained as follow. First the usual LA/AIDS budget share demand functions using a Stone price index to avoid non-linearity are developed. To represent a system of demand functions, a set of three theoretical restrictions – homogeneity, symmetry and adding-up – are imposed and tested. To capture the influence of demographic and food safety factors, like BSE and CWD incidence, on meat consumption, Pollak and Wales (1981) proposed linking the demographic characteristics to the AIDS model as translation tools (Jones et al, 2003). Therefore, following the demographic translating method, demographic data and BSE, CWD media information (collected from tracking print media coverage of the two animal diseases from a national newspaper, in this case from the “Globe and Mail (Canada)”) are incorporated into the LA/AIDS model and the extended demand function can be expressed as follows: , where, is the budget (expenditure) share of the ith good; is the intercept net of demographic and food safety effects; is the demographic variables and food safety media indices; are the prices deflated by CPI; and ln Y is the log of total expenditure. In this study, eight types of meats – Bison, Elk, Venison, Beef, Pork, Chicken, Turkey, and Seafood – are used and from the adding up property of the LA/AIDS demand system, one expenditure share equation (seafood) from the system is excluded. Using full information maximum likelihood (FIML) estimation procedures, the above equations are estimated simultaneously. This study used annually aggregated Homescan™ household data (ACNielsen) for the years 2002 to 2008. The price data for bison, elk and venison are from ACNielsen Market TRack. Bison prices are computed as an average of retail prices of ground bison, roast bison and steak bison. For elk and venison, the annual average retail price of all types of elk and venison are calculated. Since there are no regional price indices for bison, elk and venison, the prices used for bison, elk and venison are the same for all provinces. Chicken and turkey regional retail prices are obtained through Agriculture and Agri-food Canada (AAFC) web page. For beef and pork, national retail prices are derived as follows: the national retail prices of beef are computed as an average of prices of round steak, sirloin steak, prime rib roast, blade roast, stewing beef and ground beef. The national retail price of pork is approximated by the price of pork chops. In the next step, regional beef and pork retail prices are calculated by multiplying national retail prices of beef and pork with the regional price indices of beef and pork divided by the national price indices of beef and pork. The media coverage for BSE and CWD cases were collected from the Globe and Mail (Canada) through Factiva.com’s search criteria. For the household demographic data, dummy variables are used for the data defined in index numbers, which are language, region, household size, age of children, education of household head, and location (rural or urban). The midpoints are used for demographic data defined in value terms, such as the age of household head and household income. Because households who consumed the above exotic meats are not the same in all years, an unbalanced panel data for 2002 to 2008 was used in this study. Among total households (9304) surveyed by ACNielsen in 2006, only 7.4 percent consumed these exotic meats. If the sample for 2006 in the analysis is compared to the total population from 2006 census of Canada, for example, the sample contains 66 (34) percent of households with size less than or equal to 2 members (greater than or equal to 3 members), while 2006 census population exhibits a breakdown of 60 (40) percent. Similarly, for the sample, data on children’s age and being present in the family, the sample contains 30 (71) percent for the presence of children 18 years and under (none), which is a close representation of the 2006 census population of 26 (74) percent. Meat expenditure shares for bison (3%), elk (2%) and venison (2%) are relatively low as compared to those of beef (35%), pork (18%), chicken (26%), turkey (6%) and seafood (7%) in the 2006 surveyed data. However, a significant increasing trend in expenditure share from 2002 to 2008 can be observed for bison (2 to 4 percent), elk and venison (0.6 to 3 percent). A decreasing trend in beef (41 to 34 percent), pork (19 to 17 percent), chicken (27 to 25 percent), and a slightly increasing trend for turkey (4 to 5 percent) and seafood (5 to 8 percent) are also observed for the sampled households. Therefore, exotic meat types are tending to be consumed slightly more replacing domestic meat types beef, pork and chicken. From the results, it can be observed that all the own price elasticities have negative signs and expenditure elasticities have positive signs as a priori expectation showing the right curvature of underlying expenditure function and downward sloping demand (expenditure share) functions. Since the expenditure elasticities for beef, pork and chicken are slightly greater than 1, we may say that they are normal good to slightly luxury goods. Seafood, turkey, bison, elk and venison are normal to necessity goods. More interesting factors can be figured out from the results of demographic characteristics and BSE/CWD media indices. For example, 1) preferences for meat types are different across provinces; 2) pork and turkey are preferred by more English speaking people than French speaking people; 3) the larger the household size, the more elk and venison will be consumed; 4) the households with children under 18 consume less beef and more chicken; 5) elk, venison and turkey consumption increase and beef consumption decreases as income increases; 6) higher educated household heads choose more chicken, seafood and less elk, venison and pork; 7) urban people prefer more pork and less seafood; and 8) consumption will switch from beef/bison to elk/venison/turkey at the time of BSE reports and vice versa for CWD reports.

Details

PDF

Statistics

from
to
Export
Download Full History