Evaluating the External Validity of Experimental Auctions: The Case of Hydroponic Lettuce



Journal Title

Journal ISSN

Volume Title



Agribusinesses have been investigating alternative methods for food and agriculture production as a way to differentiate their product to consumers. Although experimental auctions have increasingly become a popular tool for gathering information about consumers and their valuations of differentiated products, little is known about their valuations for vegetables that are grown using different methods and few studies have simultaneously studied external validity and the influence of outside prices. This study investigates consumers? valuations of different agricultural production methods and provides a unique method for studying whether consumers? purchasing behavior reflects their willingness to pay. Additionally, it provides insight into the relationships between consumers? valuations and their prestige-seeking behavior and health-consciousness.

A Vickrey 2nd price auction was conducted and immediately followed by the introduction of an on-site secondary market that used induced value theory and the retail prices of the auction products in surrounding stores in the Bryan-College Station area of Texas. Several econometric models were thereafter estimated using data collected from the experiment; however a random parameters tobit model proved to be most appropriate due to the heterogeneous nature of the data. Results indicate that consumers express deep discounts for red colored lettuce. While tasting the products did not have an impact on valuations, consumers did express significant premiums for organic lettuce after they learned about hydroponic lettuce production and the growing methods of the products were revealed. The same can be said for valuations of mixed lettuce, which was hydroponically grown.

Consumers were also categorized using three applications of Latent Class Analysis and responses to scale-style questions about health-consciousness and prestige-seeking buying behavior. The willingness to pay for lettuce was estimated using a random parameters tobit model for each latent class in each application.

In addition to advancing the understanding of consumers? valuations of horticultural production methods, this study contributed to the external validity of experimental auctions. By using an on-site secondary market, it was discovered that consumer surplus and the relative importance of the compensation fee affect an individual?s behavior in the experiment setting.