Willingness to Pay for Low Water Footprint Food Choices During Drought

Abstract

Willingness to Pay for Low Water Footprint Food Choices During Drought Hannah Krovetz, Rebecca Taylor, and Sofia B. Villas-Boas ∗ December 21, 2016 COMMENTS WELCOME Abstract In the context of recent California drought years, we investigate empirically whether consumers are willing to pay for more efficient water usage in the production of four California agricultural products. We implement an Internet survey choice experiment for avocados, almonds, lettuce, and tomatoes to elicit consumer valuation for water efficiency via revealed choices. We estimate a model of consumer choices where a prod- uct is defined as a bundle of three attributes: price, production method (conventional or organic), and water usage (average or efficient). Varying the attribute space pre- sented to consumers in the experimental choice design gives us the data variation to estimate a discrete choice model based on a conditional Logit specification and a ran- dom coefficient mixed Logit specification. We find that on average consumers have a significant positive marginal utility towards water-efficiency and estimate that there is an implied positive willingness to pay (WTP) of about 12 cents per gallon of water saved on average. Moreover, when informing consumers about the drought severity, this increases the WTP for low-water options, but not significantly. We find that there is heterogeneity in the WTP along respondents’ education, race, and also with respect to stated environmental concern. Our findings have policy implications in that they suggest there to be a market based potential to nudge consumers who want to decrease their water footprint and follow a more sustainable diet, namely by revealing informa- tion on the product’s water footprint in a form of a label. Simulations of removing low water labels from the choice set attributes imply significant consumer surplus losses, especially for the more educated, white, and more environmentally concerned respon- dents. Keywords: Choice experiments, discrete choice model, random coefficient mixed Logit, Willingness to pay, Labels, Information, Water footprint, Food production, Cal- ifornia agriculture, Counterfactual policy simulations. JEL Classification: M30, Q18, Q25, Q21, C12, C24. Author Affiliations: Krovetz is an Environmental Economics and Policy alumnus from the University of California, Berkeley, Taylor is a PhD student in Agricultural and Resource Economics (ARE) at the University of California, Berkeley, Villas-Boas is a Professor of ARE at the University of California, Berkeley. Corresponding author: sberto@berkeley.edu. We thank the College of Natural Resources Sponsored Program for Undergraduate Research for financial support in the survey implementation.