Relatore: prof. Giovanni Millo, DEAMS

Abstract

The quality of wine can be measured in many ways. Generally, wine evaluation is based upon sensorial analysis of wine experts who assign subjective ratings. Another way of measuring the quality of wine is through market prices. For entire vintages of one wine quality, sometimes the governing body (consortium) of the local wine makers issues a quality assessment, e.g. every year the Brunello Consortium issues a one to five star evaluation of the current vintage.

The assessment of a vintage is generally needed much earlier than actual tasting becomes possible, both for reasons of marketing and of pricing; especially so in the case of the finest wines, whose future contracts are regularly traded on international markets. This time lag is particularly long in the case of the finest reds. Every vintage of the Brunello di Montalcino, for example, ages for a total of five years in barriques and then in bottles before being put on the market.

The climate during the year, through the growing season in the Winter, the ripening of the grapes in the Summer and the harvesting during the Fall, is well known to influence a vintage's quality. Ashenfelter has famously employed data on rainfall and sunshine through the seasons to predict the quality of current crops. More recently, these models have been employed to assess the effects of climate change on future crops and vintages.

We will review some well-known approaches to the econometric evaluation and prediction of fine wines' quality, replicating some results of the original applications on Bordeaux wines, before setting out to present our own original application of an Ashenfelter-style model to Brunello di Montalcino, including an extension to panel data methods by considering the ratings of different producers.