文摘
This dissertation examines empirically the values that consumers place on attributes associated with food products that are difficult to verify in the absence of product labeling,including organic,environmental and country of origin attributes. The first essay addresses the impact of product labeling and information treatments on a consumer's willingness to pay for a processed and packaged food through data collected from an online choice experiment. Participants completed five choice experiments in which alternatives varied across four unique attributes: price,country of origin,organic content and environmental friendliness. Participants also faced varying levels of information regarding the non-price attributes. Multinomial and random parameters logit models are used to measure the underlying random utility model preference parameters,which are then used to derive the marginal willingness to pay for changes in attribute levels. It is found that heterogeneity in preferences is significant amongst consumers and that the presence of information regarding product attributes affects both the mean and standard deviation parameters for product attributes. Furthermore,country of origin labeling drives both the highest and lowest WTP estimates,followed by organic content and environmental labeling. The second essay takes a broader look at WTP estimates for organic foods by use of meta-analysis,where heterogeneity of values observed in the literature is explained by both factual and methodological sources. A total of 29 papers yields 132 observations for analysis and a meta-regression is estimated using percentage premium as the dependent variable and both product and study characteristics as independent variables. Factual heterogeneity explains 65% of the explained variation in percentage premium and includes variables describing the food type under study,year of the sample and sample representativeness. Methodological heterogeneity explains the remaining 35% of explained variation and includes variables describing the data elicitation method and study methodology. It is found that studies investigating organic fruits and organic foods that are sourced from animals have higher premiums,studies using contingent valuation methods have higher premiums,and that the degree of sample representativeness of a study has significant effects on premium estimates.