Should Port classification be re-thought?

[JacobH]

Port is both an over- and under-regulated wine. Over-regulated in that oddities and innovative products usually has to be sold as a basic ‟ruby” because the categories cannot be adjusted for them and may even have to be labelled with a euphemism (e.g. ‟Reserva Especial Branco”) rather than a proper description of what is in the bottle. Under-regulated in that the supposed kite-mark of excellence, that of ‟Vintage”, has consistently not been met with certain shippers.

What interests me, however, is whether the basic categories need further adjustment to make them properly fit for their current use since, to me, the current system seems to fit more neatly with what the market would have been like in the 1960s than in the 2010s. In particular I think of the following problems:

Vintage v. Single Quinta Vintage Port
Since the 1950s, Single Quinta Vintage Port has developed into one of the most significant products in a shipper’s portfolio. As an affordable vintage Port it has great attractions to many. Equally, for me at least, there is an additional attraction to know that the Port has come from a particular vineyard; the romantic association between wine and terroir that is common elsewhere in the wine world. However, Single Quinta Vintage Port is essentially unregulated when it comes to the ‟single-quinta” part. A Port with Quinta da X on the bottle could come from anywhere in the Douro. There is a subtle acknowledgement by some shippers of this: I believe that Quinta do Noval sells single-quinta wine as ‟Quinta do Noval” and other wines as just ‟Noval” and that Grahams did a similar thing with Quinta dos Malvedos. But is this enough to properly inform consumers and prevent abuse? I am not sure, especially when one considers that the major shippers have continued to expand their larger Quintas so that, for instance, the modern day Quinta de Vargellas is significantly larger to what was regarded as Vargelas in the 18th Century.

Would an additional guarantee of origin be helpful for Port? Perhaps so, if no more than the equivalent to a ‟mis en bouteille” statement on French wines.

Late-Bottled Vintage
The essential problem with this category is that it defines two completely separate products: a ruby Port from a specific harvest, and a vintage Port that has been matured in wood for extra time. One is ideal for drinking immediately, whilst the over will mature and improve. And, of course, one needs to have its sediment removed whilst the over does not. These are poles apart and the consumer has no idea when buying a bottle which he is likely to get unless he has the chance to inspect and look for clues (such as whether the Port is filtered or not). Particularly when ordering from a catalogue or on a wine-list, can the consumer tell which sort of Port he will get when buying LBV? I am not so sure. Should the ‟modern” LBV therefore become a ‟Vintage Ruby” whilst an LBV be preserved for near-VP quality wines, simply bottled a few years later than usual? Perhaps so.

Tawny Ports
As Tom regularly points out, a Tawny-Port with an indication of a age, unexpectedly shows the average age of the bottle rather than the minimum age. This puts Port at odds with the standard in other drinks (e.g. Cognac or Scotch) which go for the minimum. Perhaps that should be changed. Equally, I wonder if the ages make much sense. For instance, how much 30-year-old tawny Port is sold? I would have thought most consumers (unless they are buying Niepoort which, I think, stops at a 30-year-old) would upgrade straight to the 40-year-old. Could that category be scrapped? Or, more radically, would there therefore be any detriment in giving shippers a freedom to make a tawny with any indication of age (as occurs, for instance, with Scotch) so that, for instance, a particularly good 15-year-old tawny could be sold, or even, if there were a market, a 100-year-old tawny?

Anyway, these are just some thoughts on the current classification system and the changes I might make as an arm-chair regulator. Any other thoughts or comments?

[DRT]

Some interesting thoughts in there, Jacob!

On the VP v SQVP part, my view is that they are both essentially the same style of wine. The fact that one is (or should be) made from a single estate doesn't change what the product is but merely, in most instances, limits the complexity that can be obtained by the winemaker due to having a restricted supply of component wines. On the subject of ensuring that an SQVP does only come from one quinta, I am not sure how other wine-making regions regulate and control that issue but having seen the Douro at work in the harvest I can't imagine that any commercially viable police force could enforce that so I think trust must be part of the game here. At the end of the day, SQVP is VP, so does it really matter if a few batches of grapes from another property end up in the blend?

I agree that the LBV problem needs to be addressed as the current situation is confusing. Unfortunately, I think the consequence of sorting this out would result in unfiltered LBVs becoming more expensive than they are today. Whilst that seem like a bad thing to us it might not be viewed in the same way by the producers. Take Warre "Bottle Matured" LBV, for instance, and compare it with Warre Quinta da Cavadinha SQVP. Which one do you think is more expensive to produce and get to market?

Tawny? I think it's already confusing enough - we don't need more categories!

Derek

[Glenn E.]

Tawny Ports
As Tom regularly points out, a Tawny-Port with an indication of a age, unexpectedly shows the average age of the bottle rather than the minimum age. This puts Port at odds with the standard in other drinks (e.g. Cognac or Scotch) which go for the minimum. Perhaps that should be changed. Equally, I wonder if the ages make much sense. For instance, how much 30-year-old tawny Port is sold? I would have thought most consumers (unless they are buying Niepoort which, I think, stops at a 30-year-old) would upgrade straight to the 40-year-old. Could that category be scrapped? Or, more radically, would there therefore be any detriment in giving shippers a freedom to make a tawny with any indication of age (as occurs, for instance, with Scotch) so that, for instance, a particularly good 15-year-old tawny could be sold, or even, if there were a market, a 100-year-old tawny?

You are furthering a common misconception! A Tawny with an Indication of Age (hereafter TWAIOA) does not indicate anything about the actual age of the Port on the bottle. The indicated ages are target profiles, not actual ages, so the indication is nothing like (and is not intended to be like) the ages for Cognac or Scotch. If you want a tawny that shows its age on the bottle then you buy a Colheita.

A 20-year old Port must meet the IVP's standard for what a 20-year old Port should taste like. If you can do that by blending together 15-year old Ports, then you'll be able to save 5 years of aging. It has happened before, but is exceedingly rare. Most TWAIOA actually average more than their indicated age because it is much easier to get them approved that way. Fernando once told me that Kopke's 10-year old is actually more like 14 years old on average and that their 20-year old is much closer to 25 years old.

Some houses produce 10/20/30 (Niepoort). Some produce 10/20/40 (Noval). Some only produce 10/20 (Ferreira, I think). One produced a 50 that it cleverly managed to sneak past the IVDP (Porto Rocha's The Glorious 50th Anniversary). I think all four are valid products, though I only really drink 20s and 40s. To me, 30s provide a less expensive alternative that's probably 90% of the quality of a 40. I just don't bother because I've stocked up on 40s during Roy's FTLOP buying opportunities.

[DRT]

Tawny Ports
Some houses produce 10/20/30 (Niepoort). Some produce 10/20/40 (Noval). Some only produce 10/20 (Ferreira, I think). One produced a 50 that it cleverly managed to sneak past the IVDP (Porto Rocha's The Glorious 50th Anniversary). I think all four are valid products, though I only really drink 20s and 40s. To me, 30s provide a less expensive alternative that's probably 90% of the quality of a 40. I just don't bother because I've stocked up on 40s during Roy's FTLOP buying opportunities.

I think the main reason why most producers only make two or three of the four allowable aged tawnies is down to having sufficient stocks of old wines to make a consistent blend year after year. Perhaps the 30 yr old category is a bit redundant in the current market so an interesting way of re-using its place would be to have the allowable styles changed to 10, 20, 40 and "over 80". This doesn't exactly solve the problem of having lots of stocks of wine for the blends but the over 80 would be a very exclusive and expensive wine and would only ever be produced in small quantities.

[mosesbotbol]

I don't see any reason for a change. Further changes lead to further confusion and more change and confusion. Leave things as is.

[Glenn E.]

I think the main reason why most producers only make two or three of the four allowable aged tawnies is down to having sufficient stocks of old wines to make a consistent blend year after year.

Actually, I think that most producers make all four. The ones I named are the only ones I'm aware of who do not, but I allowed for the existence of others with my phrasing just to be safe.

I suspect that your reasoning is correct, though. It does have some interesting side effects - the Ferreira Duque de Braganca 20 Year Old is generally thought to be the best 20 Year Old on the market (or at least in the top 3). On the last Port Harvest Tour, one of our hosts theorized that this quality was because they are able to use their very old stocks in that blend instead of holding them for 30- or 40-year olds.

[uncle tom]

I don't think there are any major issues with the rubies, other than the occasional uncertainty as to whether an LBV is filtered or not - but I can't immediately come up with a neat solution to that one. LBV is such a cash cow for the producers (especially in the UK) that any change of name would almost certainly be resisted

The tawnies are more of an issue, and the 10/20/30/40 year system does seem to belong to a different age. More important is the fact that the IVDP regulation of these products appears to be in conflict with consumer protection law.

I would agree with Jacob's idea that the minimum age of the constituent wines should be shown on the label, and that the producers should be free to present their aged tawnies in much the same manner as the whisky distillers present their aged malts - to permit any age to be shown on the label, provided it is the minimum age of the blend's components.

Tom

[JacobH]

Thanks for the responses to this thread.

On the subject of ensuring that an SQVP does only come from one quinta, I am not sure how other wine-making regions regulate and control that issue but having seen the Douro at work in the harvest I can't imagine that any commercially viable police force could enforce that so I think trust must be part of the game here.

That is true, but I think it is important not to overstate the problem. For instance, the first demarcation of the Douro did not just set out where Port wines could be produced but also graded the individual quintas and set out the maximum price for their wines. If they could make a quinta-by-quinta system work in the 1760s, then I am sure they could now. Indeed, even though the current harvest is a bit confused, the times I have spoken to a wine-maker standing over a vat of wine, he or she is can usually tell me from which part of which quinta the grapes have come from: I do not think it would be an insurmountable problem, therefore, for them to ensure that single-quinta wines were single-quinta if required.

At the end of the day, SQVP is VP, so does it really matter if a few batches of grapes from another property end up in the blend?

Well, no more so than if I buy a bottle of Le Petit Cheval I'd hope it is from Cheval Blanc rather than an assortment of other producers in St Emilion. However, I think the real potential issue is not with the few bunches mixed in but with the fact that a Port producer can potentially: i) make a wine that contains very little from the Quinta whose name is on the bottle; and ii) extend his production indefinitely by buying up and incorporating surrounding properties so that the modern quinta may bare little resemblance to the historic one. That's why I thought an additional guarantee, ‟mis en bouteille”, might be required.

Unfortunately, I think the consequence of sorting this out would result in unfiltered LBVs becoming more expensive than they are today. Whilst that seem like a bad thing to us it might not be viewed in the same way by the producers.

This is an interesting point and not something I had thought of before. However, I do wonder if the pricing scheme used by the major shippers already tends to put unfiltered LBVs in a price-bracket between the filtered ones and the SQVPs (the Croft being an anomaly)? For instance, to take the Symingtons' brands, isn't a Graham's or Dow's LBV available for about £11-12 or so, the Warre's for about £15-6 and the Cavadinha at about £20? It would be interesting to see the price-differences in their whole-sale lists to see which bracket each wine is targeted at.

You are furthering a common misconception! A Tawny with an Indication of Age (hereafter TWAIOA) does not indicate anything about the actual age of the Port on the bottle. The indicated ages are target profiles, not actual ages, so the indication is nothing like (and is not intended to be like) the ages for Cognac or Scotch. If you want a tawny that shows its age on the bottle then you buy a Colheita.

Thanks for that; I suppose I always assumed that that was an additional requirement rather than the sole one. But there is is, in IVDP Regulation n.º 36/2005 Art. 5(2):

A idade mencionada no rótulo exprime o carácter do vinho no que respeita às características organolépticas conferidas pelo envelhecimento em casco, correspondentes à idade indicada.

It's a bit shocking and quite dishonest if they are approving wines which do not at least average the age on the label. For instance, should Delaforce 1985 be sold as a 1977 because its taste "corresponds to the indicated age"? If the indication is entirely unrelated to the actuality, then perhaps they ought to be called VS, VSOP and XO or similar.

One produced a 50 that it cleverly managed to sneak past the IVDP (Porto Rocha's The Glorious 50th Anniversary).

This, I think, is quite obviously mad. There's a legitimate product, and legitimate market for it, yet the bottle cannot say what is in it because of the regulation. It's not even akin to declassification of wines in other markets where the declassified wine is usually at the lower end of the market (Rioja is a good example). Quite mad.

[Glenn E.]

It's a bit shocking and quite dishonest if they are approving wines which do not at least average the age on the label. For instance, should Delaforce 1985 be sold as a 1977 because its taste "corresponds to the indicated age"? If the indication is entirely unrelated to the actuality, then perhaps they ought to be called VS, VSOP and XO or similar.

Interesting. This particular example has never bothered me before, but if I step back and examine it in relation to other similarly inaccurate advertising strategies I would have to support your suggestion. This type of fudging annoys me for other products, but for some reason never has for TWAIOA. It isn't at least 20 years old, nor does it average 20 years old. Therefore it should not be called 20 years old. Something like what Cognac uses would be more approiate, or at least less dishonest. Perhaps getting rid of one category and then calling them "aged," "old," and "very old" would work.

I don't think you can make the leap to VP (or Colheita) though, because the dates on those are the harvest dates.

[uncle tom]

Can anyone think of a good reason (beyond the dead hand of current regulation..) why a producer should not be able to take, say, some pipes of good 1985 port, blend them with some casa stock from the early seventies (of which I believe there is quite a lot, but of moderate quality) - put it in elegant bottles with Xmas-present-friendly packaging; and flog it for a tidy profit as a 25yr old Tawny?

Tom

[JacobH]

Something like what Cognac uses would be more approiate, or at least less dishonest. Perhaps getting rid of one category and then calling them "aged," "old," and "very old" would work.

Something which I hadn’t noticed before is that the regulations (Art 5.5 and and 4.6) also provide for the use of those terms with tawnies and colheitas over 10 years being able to be described as ‟Old” or ‟Velho” whilst those over 40 years being able to be described as ‟Muito Velho” or ‟Very Old”, so if the current categories are being preserved but with different titles, it would be less of a change to switch to descriptive ages.

Can anyone think of a good reason (beyond the dead hand of current regulation..) why a producer should not be able to take, say, some pipes of good 1985 port, blend them with some casa stock from the early seventies (of which I believe there is quite a lot, but of moderate quality) - put it in elegant bottles with Xmas-present-friendly packaging; and flog it for a tidy profit as a 25yr old Tawny?

I completely agree. Equally, when I visited Oscar in October he, very kindly, let us try some ‟15-year-old” tawny that he was going to use in his tawny blends. Caroline thought it was the best Port she tried during the trip. If he wanted to, is there any reason why he shouldn’t have just been able to bottle and sell it as a ”15-year-old”?

[JacobH]

The IVDP has recently (within the last couple of years) allowed white tawnies to be produced. I have been wondering about this: if a 10-year-old white tawny is defined, in current regulations, as tasting ‟like a 10-year-old white tawny” and white tawnies have not officially existed before, how does the IVDP know what, say, a 30-year-old white tawny is supposed to taste like when someone sends one into be sealed? I understand once there is an established market you can compare a new product to an existing one but how does it work with a new one?

[Glenn E.]

how does the IVDP know what, say, a 30-year-old white tawny is supposed to taste like when someone sends one into be sealed?

While the category is new, the Port is not. Some producers - not many of them, but some - have stocks of aging white Colheitas. Dalva, for instance. Those stocks could be used to blend 10, 20, 30, and 40 year old white tawnies to be used for comparison purposes.

I have no idea how they actually do it, but this is at least a possibility.