Retail-Level Hedonics and the Valuation of Milk Components John E. Lenz, Ron C. Mittelhammer, and Hongqi Shi A retail-level hedonic model for analyzing the value of milk components contained in aggregate dairy product commodities is developed and applied to household food consumption survey data. The findings support Perrin's conjecture that a hedonic approach applied at the retail level can be used to value milk components. Milk component values are found to be a function of household sociodemographic variables. The sample-weighted average implicit fat value estimate was very similar to values published in two of three previous component value analyses in the Journal, but protein values were significantly different. Key words: hedonics, milk component values, retail-level. Multiple-component milk pricing has attracted industry and academic attention since the 1940s, but only with the development of rela- tively low-cost and accurate protein testing pro- cedures during the 1960s has this pricing method become operationally feasible (Cook, Froker, and Hardin). As the name suggests, under mul- tiple-component milk pricing, the value of a given volume of milk is adjusted to reflect its component levels (fat, protein, and nonfat sol- ids). Traditional milk pricing, by contrast, val- ues only fat. Multiple-component pricing plans with separate differentials for fat and protein have recently been implemented in the Upper Midwest, Great Basin, and Middle Atlantic Federal Milk Marketing Orders'. The call for multiple-component pricing research in the re- search provisions of the 1990 Farm Bill pro- vides further evidence of renewed interest in this area. Section 116-b calls for a study of the The authors are a research associate in agricultural economics at Cornell University, professor of agricultural economics and ad- junct professor of statistics at Washington State University, and research associate at Washington State University, respectively. The authors are grateful to two anonymous reviewers for in- sightful and constructive comments on earlier versions of this manuscript, and to the Wisconsin Milk Marketing Board for pro- viding a major portion of the funding for this research. Review coordinated by Steven Buccola. I Only milk for manufacturing is subject to the multiple-compo- nent pricing provisions in these orders. Milk for fluid use is priced on volume and fat content. In June 1993, USDA recommended in- cluding multiple-component pricing provisions in the Eastern Ohio- Western Pennsylvania, Ohio Valley, and Indiana Federal Milk Marketing Orders. National All-Jersey, Inc., a national dairy farmer organization, has submitted a multiple-component pricing proposal for the Chicago Regional, Upper Midwest, and Iowa Fed- eral Orders, and USDA has responded requesting additional pricing proposals before holding a hearing. effectiveness of multiple-component pricing for balancing production, marketing, and domestic use of milk fat, and section 116-d calls for na- tional hearings concerning the adoption of mul- tiple-component pricing in Federal Milk Orders. Proponents of multiple-component pricing cite producer payment equity and pricing effi- ciency as its major benefits. Multiple-compo- nent pricing signals changing consumer preferences to processors, distributors, and re- tailers who can respond by providing consumer products with desired attributes. Differentially valuing important milk components, including fat and protein, can also encourage dairy pro- ducers to more efficiently match their milk component production to retail-level compo- nent demands implicit in the demands for dairy products (Cook; Graf; Cropp; et al.; Hillers; Kirkland and Mittelhammer). Signaling changes in consumer preferences is especially important for dairy products and raw milk, because, ex- cept for cheese (Young, Hillers, and Freeman), there have recently been shifts away from high- fat toward relatively low-fat, high-protein fluid and manufactured dairy products (Jacobsen and Walker; Graf; Cropp et al.; Hillers et al.; Fallert and Boynton; Young, Hillers, and Freeman). Perrin suggested that a product characteris- tics framework could be useful for determining retail-level milk component values, but the topic has been largely ignored. The present study was undertaken with three objectives: (a) to develop an appropriate conceptual model for analyzing retail-level milk component values with available household survey data, (b) to Amer. J. Agr. Econ. 76 (August 1994): 492-503 Copyright 1994 American Agricultural Economics Association