Are stated preferences good predictors of market behavior?

摘要

This paper analyses how consumer responses derived from contingent valuation, a stated preference approach, correspond with actual behaviour from an experiment, a revealed preference technique. It has implications for evaluating the usefulness of predictions of actual market behaviour that are based on stated preference information. Data are based on a survey of 289 consumers conducted during January 2000 in cooperation with The Food Alliance, in Portland, Oregon, USA. Consumers were asked about their hypothetical willingness to pay (WTP) and preferences for ecolabelled apples in comparison with organic and regular apples. A revealed preference model is also used, which include instrumental variables that reflect the magnitude of consumers' hypothetical WTP. Results suggested that the coefficient associated with the low range of the stated WTP variable has a negative and significant effect on the actual purchase decision and overall marginal impact of the high range of stated WTP is positive. A consumer who states that they would pay a premium for a product is more likely to actually purchase the product.