How much are we willing to pay for quality wine? A meta-analysis and meta-regression analysis

Abstract

This paper performs a meta-analysis and a meta-regression analysis on the price semi-elasticity of wine quality ratings in the context of estimating hedonic price functions. To do so, we thoroughly search for publications that have estimated hedonic functions in this market, resulting in 223 articles estimating 1595 functions, although not all of them incorporate quality ratings. Additionally, 60 aspects of the different studies are coded, including sample characteristics, incorporated attributes, and other implemented methodological choices. The meta-analysis is then performed, revealing evidence of selection and publication biases. Taking them into account, the price elasticity is found to be around 5%–6%, remaining robust to several robustness checks. The meta-regression explores the unexplained heterogeneity in the price elasticities observed in the meta-analysis. Results indicate that quality ratings are more influential in the context of red wines, and less important in the case of sparkling and organic wines. The nature of the data source matters, but generally not the type of price used except the ones obtained from auction and en primeur sales. The influence of the remaining attributes incorporated in the hedonic function is relatively small, except for other quality aspects, reputation, and environment. The econometric techniques used in the estimation do not affect the effect size. This is also the result found for the type of publication, except for studies published in books. Finally, we find that the most cited studies are those that show a smaller effect.