Consumer preferences and willingness to pay of organic sparkling mock wines: results from a two-step study.

This study was carried out to explore the role of information on alcohol content, organic labelling and packaging on consumer preferences of non-alcoholic sparkling mock wines. In a two-step study, the consumer’s expectations and overall liking of two novel brands of organic mock wines were investigated by focus groups followed by a common hedonic test combined with a choice experiment. In the first (qualitative) part of the study, twenty-six consumers were recruited in same proportions among drinkers and non-drinkers. In the second (quantitative) step, two hundred and forty consumers were assigned to two tasting groups of equal size: all were presented at least one brand of organic mock wine, while drinkers also tasted a familiar brand of low-alcohol sweet sparkling wine. A labelled discrete choice model was fit to investigate consumer preference and WTP. Five attributes were selected to describe the beverages: beverage colour (white/rosè), packaging (bottle/can), production method (organic/conventional), brand (two different brands were used, though cannot be disclosed here for privacy reasons) and price (9,10,11,12 euros/bottle). The results of the first stage (focus groups) show that both consumers groups accepted mock wine as a beverage for special occasions. Non-drinkers welcomed it, but Muslims objected the Champaignlike bottle packaging. In the second stage, results show that drinkers and non-drinkers did not differ in blind hedonic scores of organic mock wines. No gender differences were found, though younger participants’ (19-25 years) blind overall liking was significantly higher that of older participants. Interestingly, participants were not able to significantly distinguish between a non-alcoholic sparkling sweet mock-wine and a low-alcohol (7.5%) sparkling sweet wine. Results of choice modelling yielded that beverage colour is not relevant to explain consumer choice. The preferred packaging is the bottle, organic is preferred to a conventional mock wine, while both Brand X and Z are preferred to the ‘No Choice’ option. All mean and spread parameter estimates are higher when blind liking is modelled as heterogeneity around the mean of the price random parameter. Different levels of liking are associated with different price sensitivities and this influences significantly the mean WTP (which increases roughly of 50%) though the spread of the WTP distribution is higher, and the right tail is longer. However, the model that includes blind liking scores is a better representation of consumer (heterogeneous) preferences.