Model Portfolio – 1 year results
11.01.18 – Wine Investment –
12 months ago Amphora Portfolio Management presented a model wine investment portfolio of £100,000 with a low risk bias. We are pleased to say that the portfolio rose over 13.5%, whilst the Liv-ex indices rose between 5% and 10%. As we shall see, investing in fine wine is becoming more like investing in stocks and shares as the market expands, a feature attested to in many recent Liv-ex reports.
This means that all investors have to be very careful as to what exactly they buy. You can’t just pick up a few cases and expect all to be well. The market is now much too broad for that. 30 years ago the range of availability was so narrow that virtually anything you bought went up in value. Not now, as is patently obvious from the evidence which follows.
When assembling the portfolio a year ago we took the view that an overweight position in Bordeaux was merited as the recovery in prices from mid 2015 would lead to an improvement in confidence in the sector, and whilst that has been true for the Super Seconds and the Second wines on the Left Bank, the First Growths have continued to drag their feet. The Lafite 2010 comfortably beat the Fine Wine 100 by rising 12.5%. Margaux 2012 was not far behind, but Mouton 2009 and Haut Brion 2006 barely rose at all.
The Super Second exposure fared much better, with Mission Haut Brion 2008 and Ducru Beaucaillou 2011 both up over 14%, but the portfolio stars of the Left Bank were the Second wines, which in aggregate rose over 21%. The fact is, of course, that the net exposure to the Second Wines was much smaller because as a sector it is itself quite limited, so the overall portfolio did not benefit as much as it would have done had, say, the First Growths outperformed to the same degree.
Over on the Right Bank the portfolio performed very well, rising 18.5%, but the best and worst performers raise a very interesting question for the coming year. Amphora clients will know all about our “Pomerol anomaly” theory, and we put our money where our mouth is by including the 2008 vintages of L’Evangile, L’Eglise Clinet, and Trotanoy. The best performer in the sector was L’Evangile, rising a fabulous 40% on the year. The worst, for reasons which we must confess are completely elusive, was L’Eglise Clinet, rising only 3.5%.
We emphasise the importance of stock selection, and in endless past notes have elaborated on the importance of the algorithm in getting this right, so it is no surprise to see the algorithmic value of L’Eglise Clinet 2008 at 8.5 and L’Evangile 2008 as 4.2. The higher the number, the better the value. As we prepare the portfolio for 2018 it will come as little surprise to see us switch out of the L’Evangile and, so long as the risk profile remains intact, up the ante in L’Eglise Clinet.
Between the Rest of the World and Italy the portfolio had 20% in value at the outset, and both sectors excelled, the former up 21% and the latter up 32%. Much of the Super Tuscan performance centred around a great first half of the year for Masseto 2010, and whilst very gratifyingly it has done nothing since we sold it, part of the proceeds went into Solaia 2009 which has repaid the faith by rising 16% in the second half. A very worthwhile switch.
Elsewhere the Trophy wines selections did well, but the small exposure to Champagne and Spain had an indifferent year. Overall the fund enjoyed a 35% outperformance over the broader Liv-ex Fine Wine 1000 which seems highly creditable, particularly given the absence of Burgundy, whose contribution to that index was a rise of 19%.
We have explained our reticence in respect of Burgundy several times in the past, but as noted recently there appears to be an improvement in secondary market conditions in that sector so we are monitoring the situation very closely. We would also hold our hand up to say that we have work to do in the Champagne sector, where our high conviction investments in Dom Perignon have lagged both their own sector and the broader market.
As far as the overall market is concerned, the chart below illustrates several interesting aspects of last year’s performance, and to clarify we need to investigate the composition of the indices somewhat. The “1000” is obviously the broadest index comprising the Bordeaux 500, Bordeaux Legends 50, Burgundy 150, Champagne 50, Rhone 100, Italy 100, and Rest of the World 50. In all cases bar the Legends 50 it is the ten most recent physical vintages (not exactly ten in the Champagne sector), and as the name suggests, the Legends comprises 17 of the top Bordeaux producers’ best vintages going back to 1982.
Briefly, the Liv-ex 100 is principally Bordeaux but does have representations from Burgundy, Rhone, Champagne and Italy; the “500” is exclusively Bordeaux and comprises six sub-indices: the Fine Wine 50, Right Bank 50, Second Wine 50, Sauternes 50, Right Bank 100 and Left Bank 200; the Liv-ex 50 is the First Growths’ ten most recent physical vintages.
As we can see, it was not a brilliant year for the First Growths (the “50”) and no better, frankly, for the “100”. Remember that the “100” is broader and theoretically the Left Bank underperformance should have been offset by outperformance elsewhere. We can be reasonably sure of our ground in suggesting as much when we see this chart:
Clearly every sector, even the much-maligned Rhone, has outperformed the First Growth index, and the reason this has had little impact on boosting the “100” against the “50” is at the heart of what Amphora brings to the party. Stock selection is critical. Exposure to the market, in and of itself, is not enough. It has to be the right exposure.
The Bordeaux 500 and the Liv-ex 1000 give so much broader coverage of the favoured sub-sectors that they almost can’t help but hoover up some of the outperformance, and whilst that is why these indices have outperformed, as per the top chart, they are too broad to replicate for most, if not all, investors.
The conclusion is yet again that stock selection is vital. Over the coming days we will be settling on our selections for 2018. As always we will be reliant on the algorithm which, as the above disclosures suggest, has done us proud over the last 12 months.