Despite the continuous decrease in the total market share (about 41% according to the Liv-Ex May market report), largely caused by the uncertainties brought by COVID-19 pandemic, Bordeaux wines still seem to attract a lot of attention from both passionates and professionals. The numerous historical pricing records allowed us to build an analysis focused on a set of approximately 100* of the most traded Bordeaux Châteaux in the secondary market. Our analysis aims to provide investors with useful insights and price trends that will help them in orienting themselves in the fine wine market and in choosing what is most profitable for their businesses.
In particular, our analysis achieved four main purposes:
*While the majority of the analysed wines belong indeed to different Châteaux, sometimes more wines produced by the same Château have been taken into consideration.
The analysis covered the price variations for vintages between 2007 and 2016 and considered two figures: firstly, the release price for each vintage, i.e. the first price at which the vintage is sold during its En Primeur, and then the closing price, that is the price for each vintage calculated as an average over the months of May, June and July after 3 years from the event (between 2010 and 2019). To give an example, this is the difference between the release price of a 2016 wine and its market price in the period between May and July 2020.
The data show that the vintages and Châteaux taken into account underwent a variation between the release and closing price that fluctuated between -27% and +101%. The average price variation of each Château during the period covered by the analysis is shown in the chart below.
[Please note: move your cursor over the graphs to gather more details about them, such as the wine name or the value of gain/loss, Sharpe Ratio, etc.].
At a glance, the data show that the majority of wines displayed an increase in price, while a small minority sustained a loss.
According to our measures, the top 20 Châteaux that delivered the best performance in terms of value (i.e. the price increase since release) were, as shown in the following chart: Petit Mouton, Pavillon Rouge, Beychevelle, Pin, Fleur Petrus, Forts Latour, Clerc Milon, Clos Fourtet, Margaux, Carruades Lafite, Carmes Haut Brion, Calon Segur, Canon (Saint Emilion), Lynch Bages, Figeac, Angelus, Mouton Rothschild, Petrus, Valandraud.
Instead, if we take into consideration not only the absolute increase in price but also the relation with its variability (expressed as Standard Deviation), also known as Sharpe Ratio, the scenario changes according to the graph shown below.
The Sharpe Ratio shows the average return earned in excess of the risk-free rate per unit of volatility or total risk. The greater the value of the Sharpe Ratio, the more attractive is the risk-adjusted return. A Sharpe Ratio greater than 1 indicates that on average that particular investment has delivered excess return for every unit of risk taken.
In our set, the top 20 wines that delivered the best performance in terms of Sharpe Ratio were, as shown in the chart below: Pin, Petit Mouton, Clerc Milon, Fleur Petrus, Calon Segur, Beychevelle, Petrus, Angelus, Figeac, Pavillon Rouge, Batailley, Talbot, Brane Cantenac, Pavie, Clos Fourtet, Carmes Haut Brion, Vieux Chateaux Certan, Lynch Bages, Margaux, Clinet.
Conversely, the bottom 20 performer in terms of Sharpe Ratio (hence profitability in the short/medium term) were, as shown in the following chart: Branaire Ducru, Ausone, Clos Marquis, Malartic Lagraviere, Gloria, Pavie Decesse, Bélair Monange, Ormes Pez, Saint Pierre, Grand Puy Lacoste, Haut Brion, Grand Mayne, Lafon Rochet, Cheval Blanc, Reserve Comtesse, Sociando Mallet, Gaffeliere, Suiduirat, Yquem, Rieussec.
Even though there is still a magnitude of factors capable of influencing wine value, we have identified some variables which have an influence over the price variation within the short/medium term after the vintage has been released into the market. Two of these variables explain approximately 60% of the price variance and they take into account the relative quality of different recent vintages and the release price of a wine in relation with its critic score.
The following graph shows the correlation between the observed and predicted price variations of all the wines considered in the analysis (sample size of 700+ wines from 2007-2015 vintages), expressed as annualized CAGR (Compound Annual Growth Rate).
The next graph considers also the Sharpe Ratio, and shows the observed and predicted values for all wines. The green dots represent the observed Sharpe Ratio, while the red ones the predicted Sharpe Ratio.
Our research focused also on the single Châteaux, analysing the price variation of their vintages between 2007 and 2016. Once the data had been collected, we compared the obtained results with the price predictions produced by our model. Here are some examples:
The horizontal axis shows two dates for each point of the trend lines: the top year (for example, 2007) represents the vintage being analysed, while the bottom one (for example, 2011) indicates the closing time, that is the time at which the variation in the release price is observed or predicted.
The blue column represents the observed variation, while the orange column is the predicted variation.
Some data were not available and therefore are missing. In addition, 2016 vintage shows only the predicted price variation to be observed in 2020.
As one may observe from the analyses above, our research shows that it is possible to identify the trends in price variation for different vintages in the short and medium term.
The trends we have identified are highly correlated, among others, to variables which in turn are related to the scores of recent vintages and to the release prices.
We also built an algorithmic model which, by taking these variables as inputs, is able to predict to a significant extent the direction and the level of price changes. This model can represent a crucial breakthrough in the wine industry, as it would be a precious tool at disposal of fine wine investors, who would be able to exploit reliable data and predictions to make the most of their investments.
In addition, if investors combined this algorithm with the predictions produced by the Saturnalia Vigour Index, not only would they receive detailed information on their purchases and benefit from a helpful guideline when taking decisions in the short and/or medium term, but they would also be able to gain a competitive advantage in comparison to the rest of the market.
It is important to note that the price information provided at a given point in time can be influenced by other external (such as the current COVID-19 pandemic) and internal factors that cannot be included in our model. Therefore, it is possible, as already shown in the analysis, that the price may differ from what is predicted and expected by our models.
Ticinum Aerospace cannot be liable for any losses and damages incurred in connection with the information provided by this analysis.