Differential effects of brand, ratings and region on willingness to pay: a hedonic price approach

Abstract

How consumers respond to signals of a products quality plays an integral role in marketing strategy. While the effects of product characteristics such as brand, region of origin, and ratings have been studied extensively, research on how they interact with each other and the differential effects of these by market segment has not been studied. This paper provides a unique insight into the purchasing behavior of US wine consumers and shows how brand, region and ratings interact and how each has a different effect dependent on price segments and varietal. We attribute these differences to unobserved heterogeneity across consumers who respond differently to different signals of quality. Using data on retail wine purchases in the USA combined with third-party ratings, we used a fixed-effects model to analyze willingness to pay for wine. Our findings provide marketers with a guide to which product characteristics in terms of brand, ratings, and region are the most effective signals of quality to different consumer market segments.

Keywords:

• consumer behavior
• market segmentation
• brand effects
• region effects
• ratings effect

Acknowledgements

The authors would like to thank participants at Sonoma State University's Department of Economics Seminar Series as well as participants at the Fourth Annual Meeting of the American Association of Wine Economists at the University of California at Davis in Davis, California, for helpful comments. We would also like to thank Sonoma State University's Wine Business Institute for funding this research.

Notes

Hodgson (2008).

Carew and Florkowski (2008) use retail level data provided by a distributor to examine the affect of weather on price and ratings of wines.

Steiner (2004) uses similar Nielsen scan data to estimate hedonic price functions but does not include ratings data.

www.winespectator.com

This may be especially true for Napa and Sonoma wines if there is very little year to year variation. See Ramirez (2008) for an elaboration on this point.

Private correspondence with scan data providers reveal that they are unable to provide vintage in the product description because wine producers often re-use sku's for subsequent vintages.

The actual Napa premium at a score of 85 is 0.2%.