Bordeaux en primeur 2020: an investment opportunity?
Our Bordeaux correspondent, Colin Hay, speaks to Matthew O’Connell, head of investment at Bordeaux Index, about the investment potential of Bordeaux 2020. The conversation took place before the withdrawal of Ausone and Cheval Blanc from the St Emilion classification.
What makes for a vintage with strong investment potential at the point of release?
There really needs to be a triangulation of critical consensus, favourable comparison to other recent vintages, and release prices which give medium term potential on the secondary market. The outright quality of the vintage does not in itself matter so much; rather it is price position relative to other similar and better/worse vintages that is more significant. That said, a vintage can have mixed potential if some wines are released at significantly more attractive prices than others.
Does 2020 have all, some or none of those attributes?
It ticks many of the boxes, but not all of them. The 2019 en primeur wines clearly had strong potential almost across the board with that rare combination of high quality and attractive prices. In comparison, 2020 en primeur was much more of a mixed bag,with only a handful of wines released at prices which were better than marginally attractive.
How do you view the quality of the vintage overall and what, for you, are is distinctive characteristics? Do they have a bearing on its interest for investors?
The 2020 vintage is clearly strong, but it is relatively heterogeneous. But crucially, from an investment perspective, much of the heterogeneity was outside of the core investment wines. Yet, at the same time, there were few potentially perfect scores and an overall feeling that the vintage may, in time, come to be ranked third in the 2018-2019-2020 series. It is certainly a substantial way behind 2016. From an investment perspective there was therefore reason– particularly in the strong underlying market conditions in of 2021 so far – to be open-minded about somewhat smaller price discounts but ultimately to be disciplined where prices seemed to offer only marginal short to medium-term potential.
Does en primeur give enough incentive to the buyer and to the buyer as investor? Should it? And should it give more? Does the increasing use of analytic tools like Liv-ex’s fair value methodology actually threaten to crowd out some of the investment potential in the EP market?
There is no magic formula for an en primeur discount but ultimately if an unfinished, non physical wine with an uncertain final score is only marginally cheaper than a finished, physical wine with a final score, it is difficult to justify its addition to an investment portfolio. The investor needs to be able to anticipate an incremental positive return for an en primeur investment based on provisional scores for it to be worth the risk.
Fair value analysis of some kind has relevance for en primeur, but only as one input into establishing how attractive an en primeur discount is. It certainly shouldn’t be used to say, in effect, ‘this wine is fairly valued and therefore it’s a Buy’; that somewhat misses the point and also over-simplifies. Even as one of multiple factors, fair value needs careful handling – for example, what if a new release is below the apparent fair value, but actually other similar quality physical vintages also are? Fair value analysis also needs to be adapted to different wine dynamics (e.g. La Mission trades very differently in strong vintages; Lafite has a low correlation between scores and price, and so on). And for some wines it is important to consider a wider spread of critical scores than it is for others.
To respond specifically to your question around the impact of Fair Value analysis on en primeur, to be honest I don’t think there is much focus at all on the Liv-ex analysis itself on the merchant or consumer side (with attention more around specific recent vintages, individual wine dynamics etc.) and I think for the most part Bordeaux properties also base their release prices on a broad variety of factors – and, indeed, their own analysis of pricing.
Buyers – investors and also collectors for the most part – see the need for a discount for en primeur, so if a chateau releases a wine at prevailing market levels / fair value / a price which across key factors looks marginal, then it will inevitably struggle to sell in volume. Regardless of how the price is derived or analysed, this won’t change!
How would you rate the investment potential of 2020 vis-à-vis 2019, 2018, 2016, 2015?
In the long-term, we think of 2016 as something of a quality benchmark; as such it offers a number of advantages as an investment vintage. 2019 may yet fall into the same category. However, it is hard (and perhaps too early) to make such absolute assessments – for example, at present 2015 looks compellingly cheap in many cases. In short, the devil rests in the detail and there will always be investment potential at the level of individual wines in each of these vintages.
How is the investment market for Bordeaux en primeur evolving in your view?
There has been a clear movement towards lower volume release, particularly amongst investment grade Bordeaux wines. This can only serve to decrease the significance of en primeur in an investment context; it is also likely to risk diminishing the relative focus of investors on Bordeaux as a region. 2020 appeared to see the lower volumes used as a means to push upward prices to the point where many became marginal (i.e.: ‘can we just about sell through at this price’). If this pattern continues, it threatens to make en primeur less significant than it was for investment too. That said, it is really too early to draw conclusions on this front; but the concern is there.
What have people been buying?
Among the wines which sold well there were few genuine surprises, with significant interest in Cheval Blanc, Figeac, Margaux, Mouton and Lafite. Many of the ‘Super Seconds’ and similar wines were much more challenging price-wise in a segment which is also – ironically – rather more price sensitive.
What do you see as sure-fire wines to follow vintage-to-vintage?
At present, Cheval is proving to be a ‘model’ chateau in terms of how we would want to see en primeur volumes and pricing balanced to maintain secondary market momentum. What is interesting is that it is also reaping the rewards in the pricing of and interest in back vintages. Figeac also continues not to put a foot wrong from a quality perspective and, in our view, it also handled pricing astutely in this vintage.
What do you see as exciting new opportunities – wines on an upward curve, under-valued or under-appreciated wines?
Figeac likely falls in this category especially given the potential for ‘upgrade’ in the forthcoming St Emilion reclassification. Beyond that, this probably isn’t the right vintage to ask about underpriced wines!
Should investors be seeking to anticipate the results of the 2022 St Emilion re-classification or should they be looking elsewhere until the results are declared?
Figeac has been a market focus in the last 12 months and we have seen an uplift of 20-30% over that period. This followed a period of relative underperformance and suggests that the market was waiting to gauge the strength of its vintage-to-vintage performance rather closer to the 2022 reclassification itself. We see this as the main wine to focus on in that context and note that Canon, for example, has not received the same amount of market interest.
Analysis and reflection
Matthew O’Connell’s reflections on the en primeur campaign are very interesting. While a number of points could potentially be drawn out from them, I will highlight just four.
The first is release price dispersion. In short, we have witnessed a very considerable variance in increases in release prices across the campaign. Indeed, as I was at pains to point out in my weekly analysis of the campaign as it evolved, we have never before seen such a significant variation in release price – with the first-releasing wines typically increasing their prices (if at all) at a rate very much lower than those releasing towards the end of the campaign.
That, of course, suggests that there is relative value to be found for investor in some of the earlier releases (the only caveat being that the majority of what might be seen as ‘investment grade’ en primeur releases came in the second half of the campaign).
Secondly, and perhaps paradoxically, the campaign has not proved as highly price sensitive as one might have imagined – particularly at the top end. The first growths, despite in some cases very considerable increases in release price relative to their 2019s, have sold well – and in most cases sold out.
That is largely because of two factors: that the 2020s were released in a rising market (and one that had been rising for a number of months); and the relatively small size of the release of the leading wines (with, typically, 20-40 per cent less wine available en primeur). Both factors have conspired to make en primeur 2020 work for the leading chateaux. But, as Matthew O’Connell points out, this is a strategy that cannot endure indefinitely without killing the appetite that feeds the Bordeaux en primeur beast.
Third, and no less significantly, the heterogeneity of the 2020 vintage (when compared to, say, 2019 or 2016) has not turned out to be too much of problem for the Bordeaux investment market. The reason for this is simple: there is little doubt that almost all of the investment grade properties have made great wines in 2020 despite the heterogeneous character of the vintage on lesser terroirs.
But that is, of course, not true for the collector market. This is rather broader in the range of wines on which it is focussed and rather less concentrated on the first growths and their peers. This market has taken more of a hit relative to 2019 because of the more uneven character of the vintage.
Finally, Matthew O’Connell’s comments on the greater use of market analytic techniques – above all by producers but also by brokers, investors and collectors alike – are fascinating. Fair price accounting, if we can call it that, is potentially very helpful to both the collector and the investor (as it is for the chateaux). But there are risks associated with it too. The inferences one might draw from it, above all about the value of en primeur releases, need to be carefully considered.
Liv-ex’s increasingly influential ‘fair value methodology’ is based on the statistical analysis of the correlation between prices (current market prices for already released wines and release prices for en primeur wines) and the scores of international critics. It gives us an efficient and reliable way of comparing, in effect, the anticipated market value of a wine with a given score and the wine itself. Put simply, if the (release) price of a given wine is below the fair value curve then it represents ‘better than fair’ value and might be thought of as a good potential investment.
But current market prices for wines with, say, 10 years of bottle age and a final score from the reference critic are not the same as release prices of non-physical wines lacking a final score. To assume they are is to compare apples and pears.
Perhaps most importantly, to take the value curve as the guide to good or poor value en primeur is to fail to appreciate the risk that the investor or collector takes in purchasing a wine before it becomes physical. That risk comes principally in one of two forms: (i) that the adjusted final score for the wine places it above the (future) value curve at the point when it becomes physical; and (ii) that market trends in the intervening period have lowered what we regard as fair value (either for the wine itself or more generally).
The investor – and indeed the collector – needs some compensation for that risk (as well as for any exchange rate risk they might also be taking). And if that it is accepted, we need to apply some kind of discount to the fair value price suggested by the statistical modelling.
Interestingly, Cheval Blanc – the first property to be both explicit and public about its use of Liv-ex’s fair value methodology in deciding upon its 2020 release price – offered to the investor and collector alike precisely such a discount. That is why it worked so well. Most of the ‘Super Seconds’ did not. That is why they worked less well.
The point here is not to suggest that we are better off without fair value accounting, or similar analytic techniques for evaluating the investment potential in en primeur releases. But it is to suggest that such techniques do not provide (as yet, at least) a sure-and-fast guide to current value, let alone future appreciation. They are no substitute to, and can never be allowed to replace in some algorithmic manner, the fine-grained analysis of releases on a case-by-case basis.