Thanksgiving food (or drink?) for thought: What you appreciate appreciates—intriguing idea to apply to fine wine investing.
We can get so caught up in the math, the science, and the data in analyzing anything these days—so much so that we might have moved 20 steps away from the basics, and forgot what actually makes sense. This happens quite frequently in the investing world, even when it comes to fine wine. I recently had the great pleasure to watch a beautiful film called "A year in Burgundy," which drew me right back to the roots and basics of our wine models and inspired me to tell our wine forecast story in a way that is more connected to the fundamentals.
The film, "A year in Burgundy," follows Martine Saunier, a famous wine importer & wholesaler, as she visits Burgundy throughout the year of 2011 and spends time with the many French farmers she has known for decades. It is amazing that the way the French farmers in Burgundy passionately grow their grapes and make their wines parallels the way we approach our models: organic, holistic, and intuitive. In the past few decades, scholars and investors have been curiously exploring and evaluating aspects of wine investing, using common tools that have been available for the more traditional asset classes such as equity. Some literature noted that those tools are not for explaining the wine price appreciation and returns, which was for "future research" .
Our application of the Age-Period-Cohort (APC) models in fine wine analysis fills this void. The APC analysis has been traditionally used in epidemiology and mortality rates. In recent years, however, it has become the leading method in the Federal Reserve's annual stress testing among the financial institutions regarding their retail lending business, due to the inherent features of consumer loans . It is the first time that the APC models are applied to fine wine investing analysis, and it is a natural choice: the lifecycle of the wine (age), the environment/market condition at the time of observation (period), and the weather, the harvest, and other relevant conditions in the year at the birth of the wine (vintage) are among the most important aspects when it comes to forecasting a fine wine's price appreciation over time. By decomposing historical auction price data into these three components, we are able to apply scientific methods and mathematical tools while preserving and respecting the very nature of the wines—their life, their present environment, and their birth.
"I feel good down in the cellar because it's very calm. Since they were constructed by the monks back in the 16th century, there's a special atmosphere down here. I love it. Often I put on some classical music in the barrel room. It does me good, and I think it does the wine good. Wine is alive. From the moment you pick the grapes, till the moment you drink it, it is alive. Music calms it as it matures."—Thibault Morey, wine maker and pianist, A Year in Burgundy
This aspect is what makes our forecasts powerful and the French farmers passionate about their wines: we both respect the lifecycle and therefore the trajectory that lies within each bottle of wine. As Lalou Bize-Leroy, the "queen of Burgundy," lovingly describes the way she respects the life of her vines:
"Yesterday I spent the whole afternoon in my vines. Yes, of course, I love my vines! When they don't see me, they are unhappy. When I arrive, they are happy. I really love my vines. Vines are not well understood. You have to put yourself in their place. You have to understand why they are not doing well. You have to be part of the life of the vine, and the life of the soil too. That's all you've got to do. Of course, I believe in biodynamics. We should cut out all the herbicides, the insecticides, the fungicides, the pesticides—all the 'icides.' They sound like homicide! We should stop killing things and give them the life force instead."
And it is this "life force" that we are using in our models that gives us the foundation of the predictive power. We use historical data to extract the lifecycle that a particular type of wine (say, Burgundy) follows on average, taking into account all vintages. This lifecycle then explains the inherent (capital) appreciation of the wine, cleaned of outside circumstances and unique characteristics of each vintage. And it is a lifecycle this particular type of wine will inevitably follow—the lifecycle is inherent and unstoppable.
"Burgundy wines ought to be all different. But science now allows us to make wines that are all the same. It's a pity. They all used to have their own character, like human beings. But progress is pushing all of us to make the same kind of wine."—Dédé Porcheret
Contrary to the lifecycle, which is the internal aspect of winemaking and wine's appreciation in intrinsic value, the environment component in our model captures the external characteristics (i.e. market condition) at the time of observation (i.e. auction price). This is the "market index" that we have been referring to in our previous post . This component is the most comparable to the stock indices, although we use broader coverage in wine, and we can conveniently study any particular wine's (such as Burgundy, Lafite, Bordeaux, Italy, or California) relationship with the stock markets and other economic factors such as GDP. Similar to the other components, the environment function is also cleaned of lifecycle effects and unique vintage effects.
Coming back to the film, as winemaker Dédé Porcheret points out, one challenge for the great winemakers in Burgundy comes from the market condition and demands in the current environment. The narrator of the film also poses this question: "How much will the standard tastes of the international market force Burgundy to change? How will the wine makers survive, if they insist on allowing each vintage to be its own unique self? ... The future of Burgundy depends on the taste of connoisseurs…"
Although we cannot predict the future dynamics between consumers and winemakers, we can choose economic factors and stock markets that have been relevant to the wine markets in the past, and draw insights from the forecasts of these factors, which will in turn contribute to our forecast for the wines.
"It's quite something to drink a wine you made yourself 35 years ago."
A wine's vintage is the year the wine was born. In our APC modeling, this is the third component and it captures the value of each specific vintage relative to the lifecycle function, the latter of which shows the appreciation of an average vintage over time. Modeling by vintage is a powerful idea because it is an essential feature of wines, and it reveals what seemingly unrelated topics could have in common at the core, such as fine wine investing and retail lending in banking, where each loan is a "vintage." In winemaking, each vintage is affected by that year's condition such as weather and harvest, and the unique blend of yeasts. Lalou Bize-Leroy talks about the yeast and the birth of the wine:
"Each vine has its own yeast. Yeast gives a wine its identity because it is part of the grape. It's in the earth and the air too… It's yeast that's the winemaker—not us. It's yeast that breaks down the grapes and changes sugar into alcohol. The alcohol kills off the yeast, but it's the yeast that gave birth to the wine. That's a good lesson to learn."
Indeed, there is much to learn from the wines, winemaking, and the French farmers, who bring life to their creations. Comparatively, APC modeling is an intuitive and holistic approach that incorporates the components essential to providing more accurate forecasts. I believe that bringing APC modeling into fine wine investing pays tribute to the life of the wines, and will yield profound insights for investors.
Sisi Liang currently works as a risk consultant, focusing on regulatory stress test modeling, and financial risk practice. Before her current role, Sisi has experience in Equity Research covering metals mining (production and commodities), energy, and the media sectors. Sisi holds a BA in English and Mathematics double major, with Honors, from Washington and Lee University, and an MS in Quantitative Finance. Her passions for forecast modeling, the poetic and commodity elements in wine, yoga, along with her curiosity in the interconnectedness in diverse fields have drawn her to the collaboration of auctionforecast.com.
 Lee W. Sanning, Sherrill Shaffer and Jo Marie Sharratt. "Bordeaux Wine as a Financial Investment," Journal of Wine Economics, Vol 3, Number 1, Spring 2008, Pages 51-71.
 Joseph L. Breeden and Jose Canals-Cerda. "Consumer risk appetite, the credit cycle, and the housing bubble. In Credit Scoring and Credit Control XIV Conference, Edinburgh, 2015.
 Sisi Liang and Joseph L. Breeden. "Shipwreck of the Chinese Stock Market May Lead to Sunken Treasure in the Fine Wine Market," http://auctionforecast.com/Blogs/Details/6 , Aug 20th, 2015.
 Martine Saunier, David Kennard. A Year in Burgundy, 2013.
Yoga photo credit: Elyza Bleau