Fine wine investment: Holding the fort

The series of articles depicting the performance since the market rallied at the end of November 2015 of wines both favoured and otherwise by the algorithm has proved very popular, and before moving on to other things this week we will answer a question that has come up regularly, concerning the Second labels of the first growths.

This sector is interesting because pricing doesn’t appear to follow the same dynamics as other areas of the market, whereby, for example, a first growth or ‘Super Second’ ‘off-vintage’ will cost considerably less than an ‘on-vintage’, often by a factor. Most people who claim to know anything about wine will be interested as to vintage, and this isn’t just wine-snobbery, although much else is!

The quality of a vintage is considered a key determinant in the ultimate quality of a wine, and you will occasionally find wines with superior scores costing less than other wines from the same producer simply because they come from inferior vintages.

Second label wines, however, are much more homogenous as far as pricing is concerned. Nine of the last 10 physical wines from Petit Mouton range in price only from £2,100 to £2,300, the 10^th being something of an outlier: the 2007 costs £2,400. The Robert Parker scores span 93 to 87, the pricey 2007 being a lowly 88.

In such a world could the algorithm have helped engineer outperformance, at all?

Leaving aside Forts de Latour for a reason we will discuss later, let us first have a look at Petit Mouton. You will recall that we input the scores from 1 December 2015 into the current algorithmic variables, to see what would have been thrown up as top buys and sells at that date, and compare with actual performance since then.

Top buys for Petit Mouton were 2009, 2010, and 2014 (then available en primeur), while the top sells were 2001, 2003, and 2007. Since then the average rise for the ‘buys’ has been 97%, while the average for the sells has been 46%. Of course 46% looks a perfectly decent performance, but that is not the point at issue here. What we are aiming for is maximum potential return, and the ‘buys’ have outperformed the ‘sells’ by an astonishing 110%!

Even more interesting in this case is that one of the sells would have been the above mentioned ‘outlier’ from 2007. At its unusually high price of £2,400, it has risen a substantial 105%, and yet even with this kicker the ‘sells group’ still underperformed. Imagine the underperformance had it not been quite such an outlier.

Carruades de Lafite tells a similar story, even as to the ‘buy’ vintages. In this case the average rise of the 2009, 2010, and 2014 is an even more extravagant 113%. The ‘sells’ differ in that the bottom three were 2002, 2004 and (again) 2007, and their average rise was 50%. Not to be sniffed at, perhaps, but an even greater underperformance of 126%.

We’re getting used to reappearing vintages in this study and Pavillon Rouge doesn’t disappoint, even if the outperformance does, a paltry 40%. The trio of ‘buys’ is the same, but for Pavillon Rouge on the ‘sell’ side the ubiquitous 2007 is now joined by the 2000 and 2002.

I think we can say with some degree of confidence, as a result of the analysis we have disclosed in recent weeks, that the algorithm would have been very useful identifying exactly which wines to buy at the start of the rally. Not only that, but in most cases the degree of outperformance is really significant, which suggests that it has considerable value for the fine wine investor.

I mentioned that we would return to Forts de Latour. We at Amphora think the market finds the Latour stable a bit of a challenge, but it is replete with opportunity nonetheless. Latour pulled out of the en primeur game after the 2011 release, and we are happy to discuss this in detail should anyone be particularly interested. Suffice for the moment to say that the château stores the wine by itself, releasing to the market when it thinks the wine is ready for consumption.

What it also does is release stock later in the drinking window, seemingly at random, at a premium to the going-rate, a practice we have reflected upon critically at various times in the past. Investing in these ex chateau releases doesn’t seem to have been a winning strategy thus far, and they have simply helped cloud the investment picture, in our view.

In light of this it may not come as too much of a surprise that over the last 30 months or so the algorithmic ‘sells’ for Forts de Latour have done just as well as the algorithmic ‘buys’, but we would contend that far from invalidating the analysis, what that does is reinforce the current opportunities. Ultimately, value will out.

It is very unusual for a Second label to score in the high 90s, the typical range being 88-92 or so. In 2010, undeniably a fabulous vintage, Forts blew the lights out with a 97, the highest score ever achieved by any First Growth producer with their Second wine. This wine is available more inexpensively than the 2000, 2001, 2003, 2005 and 2008, all inferior efforts.

Latour is undoubtedly doing something with its Second wine which the other First Growths are struggling to match. We will be looking more into this in the coming weeks, but for the moment we recommend you buy Forts de Latour 2010.

Philip Staveley is head of research at Amphora Portfolio Management. After a career in the City running emerging markets businesses for such investment banks as Merrill Lynch and Deutsche Bank he now heads up the fine wine investment research proposition with Amphora.