Thursday, May 26, 2011

Alcohol in wine: increasing in every country

My bloodstream's alcohol is also underreported
It's not news that wine is getting higher in alcohol. What is news -- from a research paper issued yesterday -- is how systematic and universal the trend is.

Moreover, EVERY country reports less alcohol on the label than actually exists in the bottle. It's interesting, at a time when people are looking for bold wines, that wineries worldwide apparently believe slightly lower alcohol percentages on the label will help the wine sell better.

And for as much heat as California wineries take over alcohol, the growth in alcohol percentage in the US over the last two decades is actually the lowest of any major winegrowing country in the world.

The research paper -- published by the American Association of Wine Economists and cowritten by four researchers from UC Davis* -- used data since 1992 from the Liquor Control Board of Ontario, which tests the actual alcohol content of every wine it imports.

* Authors: from UC Davis, professors Julian Alston and Jim Lapsley, PhD candidate Kate Fuller and research associate Kabir Tumber. From the Liquor Control Board of Ontario, VP George Soleas. 

There's a lot of great data in this study, which I would link to if I could. But I can't, so I'll summarize some of the most interesting points below.

* Alcohol is rising in wine in every country in the world much faster than can be explained by global warming.

* There are a few regions in the world that counter this trend, with alcohol actually falling since 1992: Burgundy (reds and whites), Oregon (reds only), Piedmont (reds only), Washington (reds only). Oregon reds have actually dropped by the most, although I wonder if that indicates a change in grape varieties; the paper doesn't address it. Piedmont's and Washington's drops were tiny, but that still sets them apart from the rest of the world. And, hurray for Burgundy!

* The regions that have added the most alcohol are fairly obscure: British Columbia and US Other (not the 3 West Coast states) for red wines; British Columbia and Canada Other for whites. This makes sense, as obscure regions must be pursuing ripeness.

* Among major regions, Languedoc and the Rhone have added the most alcohol to red wines. For whites, it's the Rhone, Languedoc and Piedmont.

* By country, the fastest alcohol growth rate in the world for red wines is in Chile, by a fairly large margin over Spain and South Africa. The US started in 1992 with the highest average alcohol of any major winegrowing country (13.5%) but the growth rate since then is actually significantly less than the rest of the world; France has added alcohol at nearly three times the US rate.

* By country, the fastest alcohol growth rate in the world for white wines: Would you believe Canada? New Zealand is second, followed by Chile and Italy. Once again, the US had the highest average alcohol to start with in 1992 (13.4%) but since then our growth rate is the smallest of any major winegrowing country in the world; Italy's alcohol growth rate is more than five times ours.

* Before you ask, one flaw of this study is that 1992 is the last year for which they published the average alcohol. For the rest, it's all annual percentage change, trend growth rate, regression analysis, logarithmic stuff, etc. Academics ain't journalists. Be glad I'm filtering.

* Every country underreports alcohol for both red and white wines. Most grievous for red wines: Spain, just a touch less honest than the U.S. For white wines, the U.S. and Chile are equally inaccurate, towering above (below?) the rest of the world.

* Hurray for Portugal: that country's wines are the closest to having the alcohol reported accurately for both red and white wines. Just another of many reasons to love Portuguese wines ... New Zealand finished a close second in honesty for reds, not as close and tied with South Africa for second in honesty among whites. Wines, that is.

* The authors use statistics I don't understand to extrapolate the "desired" alcohol level from each country; in other words, what wineries think we want, based on what the actual alcohol level is and what they report it as.
The "desired" alcohol levels seem really low to me: Red wines worldwide, 12.98%. Red wines in the US, 13.21%. I think this might be a flaw in the study, and possibly a result of mixing corporate $7 wines with aficionado $30 bottles, but I'm not smart enough to say for sure.

* For average alcohol of all red wines since 1992, Australia is the highest at 14%, just a wallaby's nose hair ahead of the US at 13.99%. Canada's is the lowest at 12.8%; France is second at 13.1%.

* Here's the real hotspot of US wine, and it's rarely talked about. For all white wines since 1992, US wine is by far the highest at 13.66% alcohol; Chile at 13.45% is a distant second. Those delightful Portuguese whites that people don't drink enough of have the lowest alcohol, 12.33%, followed by Italy, 12.39%.

So remember, when it comes to American heat, it's not the Zinfandel that stands out; it's the Chardonnay.


WineDIneDIvas said...

Thank you for the great article Blake! The increased alcohol level is the culprit, and we thought that we are getting less tolerant when tasting multiple wines...
Judit & Corina

Christian Miller said...

Fascinating stuff, I'm going to be looking this up. But I have to add one caveat right off the bat - the study is of wine submitted to the LCBO, i.e. intended to be marketd in Canada, no? That's a broad selection, but not a sample representative of wine produced in general, or even necessarily for home or major markets.

Jerry Murray said...

In regards to Oregon:
I think it would be useful to note that there is likely an increase in the variability in alcohol levels from vintage to vintage. It seems in the last decade we are frequently seeing the "coolest vintage" in memory followed by the "hotest vintage" in memory and the extremes could well be gettin further and further apart.
I also have to point to Oregon winemakers learning curves. In 2003 a winemaker unwilling to add water to a fermentation was common. By 2009 a winemaker unwilling to do so would of been an outlier. We have learned to make less alcoholic wines.

Larry Brooks said...

You know I think this whole rising alcohol phenomena should be put in perspective. I believe that the amounts of lacohol we're talking about are not that significant in terms of effect and rarely have significant effects on flavor. The difference in amount of alcohol in a glassof 14.5% vs. 13.5% is equivalent to 1/10th of an ounce of 90 proof booze. If you drank an entire bottle of that same wine it would only be equivalent to 1/2 and ounce of 90 proof. Over the course of an evening that's just not significant.

Jim Klein said...

I don’t believe that most winemakers are enraptured with alcohol. (Craft brewers yes, but not winemakers. And don’t even get me started on the over use of hops vs. the over use of oak) A winemaker’s job is to make the best tasting beverage possible given the raw materials, environment and technology available. Lots of factors are contributing to the raise of wine alcohol, and many of these occur before the grapes hit the crush pad. Here’s my short list as to why alcohol is higher today. Call it my road to the hell of high alcohol wine:

- Virus free plant material
- Drip irrigation and fertigation
-Rootstocks that need to be kept vegetative late into the season
- Vertical trellis with highly exposed fruit
- Basel Leaf removal
- Higher degree of monitoring plant macro and micro nutrients and supplying those needs by either organic or synthetic means
- An obsession with visual fruit appearance, especially lignified stems and seeds
- Colder fermentation temperatures, especially for whites
- Consumer preference for ripe fruit flavors over a more herbal profile

Anonymous said...

Link to the study (PDF): Cool insight - I look forward to digesting those wonderfully complex numbers.

W. Blake Gray said...

Anon: Thank you, thank you for the link. I couldn't get a link to a PDF to work in the main body of the text.

Jim: I don't disagree.

Larry: I do disagree because of the numbers you chose. 14.5, I'm used to it. I just got a shipment of Napa Cabs and the labeled alcohol was all over 15, which means they're probably closer to 16. The difference between that and 13.5 is significant.

Jerry: Interesting point. I think there's more there, more than I can address in a quick comment.

Christian: It's a fair point, but the data is probably not available elsewhere. I think the larger issue is one I mentioned in the text: They controlled for many variables, but not for price.

Judit & Corina: Muchas gracias!

Rafa Ibarra said...

Hi. In Mexico there are wines that are over 15 or 16 % alcohol, but the wineries put in the labels "13.5%". Why? The answer is very simple: taxes.

Here our "brilliant" government decided several years ago to create a new tax called IEPS (Impuesto Especial sobre Produccion y Servicios = Production and Services Special Tax). This tax is additional to the VAT (Value Added Tax) that is 16%. But IEPS vary depending on the percentage of alcohol in the beverage (how stupid!). So:
-If the beverage contains 13.9 or less % of alcohol, IEPS is 25%.
-If the beverage contains between 14 and 20 % of alcohol, IEPS is 30%.
-If the beverage contains more than 20% of alcohol, IEPS is 50%.

SO, this is another reason to lie about percentage of alcohol in wines.

What do you think about it?

Best regards.

--Rafa Ibarra
El mundo de Rafa Ibarra

Urs E. Gattiker -- @ComMetrics said...

Great post. I am surprised that you did not provide a link to the paper?

That is a fascinating read and its tables at the end. Great.


Hope this is useful and thanks for sharing.

W. Blake Gray said...

Rafa: The US has a similar situation. Taxes go up at 14% alcohol, encouraging large producers to keep the labeled alcohol at 13.5%.

In theory, US wine producers are not allowed to list alcohol on the label on the wrong side of 14%; i.e., if the label says 13.5%, the actual alcohol could be as low as 12%, but it couldn't be higher than 14%.

Anonymous said...

Can we please move on to another topic? OMG it has been beaten to death and been SO over-analyzed. Don't the wine media have anything else to talk about?

W. Blake Gray said...

Anon: As always, the wish of somebody without the courage to sign his/her name to his comment is my command! Please see today's post.

Anonymous said...

While I was not the first anonymous -- I will have to agree. My name is Andy BTW. ETOH levels, whether they are rising or falling and their degree of accuracy on the label is NOT IMPORTANT (except for taxes of course). What is important is how the wines tastes (and sells, if you are on the production / retail side). if you like or dislike how a wine tastes, then who gives a rats a$$ how what the ETOH level is? I've tasted low ETOH flaccid fruit bombs and beautifully balanced table wines at 16+% and everything in between. What makes a wine good or bad to your palate is so overwhelmingly complex that it simplely cannot be broken down to this 1 variable--thus one cannot say that the change in ETOH levels reported in this study are good or bad. Now why are EOTH levels increasing -- I agree with Jim Klein's post and I would also add commercial yeasts (newer strains with higher ETOH tolerance) and (maybe) tank fermentation as additional reasons. As an aside and to reveal my bias (or lack thereof), I make wine (home -- about 300-400 gallons per year with 2 families) and I've made a variety of wines from Northern California grapes that have suited my palate with (accurate) ETOH levels of 13.2 to 16.7%

Anonymous said...

We should be careful to state as fact the results from the recent, non-peer reviewed, article on "Too Much of a Good Thing? Causes and Consequences of Increases in Sugar Content of California Wine Grapes." While I think much of it is good work, their treatment of climate was very poor ... namely, they looked at a period of less than 30 years, which anything less has been proven to be inadequate to capture climate structure and trends, they looked an extremely warm period with little overall change (most of the change in temperature happened between the 1970s and 1990s), and they attempted to separate the issue neatly into two hypotheses which clearly can't be done. Take a look at last year's Wines and Vines special issue on rising sugar and alcohol levels and you will clearly find that there are complex multi-cause issues going on, but one where without a warmer climate growers could not hang the fruit longer, nor winemakers choose to make bigger bolder wines. So for this non-peer reviewed piece to make a bold statement that climate does not matter in the abstract/press release (not many will read the whole paper like I did), they would not see through the issues and even note that the authors discuss their data shortcomings clearly in the text (but not the abstract/press release). This appears to be another case where decent preliminary research, but with severe limitations, hits the fiber optic waves with certainty!

W. Blake Gray said...

Anon: That's a fine job of debunking a research paper I didn't refer to.

Andy (and perhaps 95% of other California winemakers): I'm sorry, but you guys are arrogant for saying that something many customers claim matters to them, in fact does not matter.

Why are wine writers, bloggers, twitterers, et al, writing about alcohol levels all the time these days? Because it DOES matter. Certainly not to everyone, but very little save the continued beneficial heat from the sun matters to everyone.

I didn't go out of my way to get attention for this post; I wrote it because there was a report, which is a news item, and at heart I still am a news guy.

This post was tweeted about quite a bit and widely read. Maybe it is the same drumbeat over and over. And maybe you should consider listening to those drums, rather than telling people they have no rhythm.

Anonymous said...

Andy again..."arrogant"...ouch, thats pretty harsh. Personally, I make wine for myself so I can do whatever I want. As you can tell by the range of ETOH in the wines I've made, ETOH is not important to me as my wines span the enitre spectrum-- I produce wines to reflect the varietal and AVA (amador barbera... 16.7%, Napa petite sirah 13.2% etc, etc, etc.) Now I do not think I am remotely qualified to be lumped in the 95% of California winemakers but hear out my point. I am not saying ETOH levels are not important to an individual -- if you like wines lower in ETOH then please, please speak with your wallet (or pocketbook). If enough people do that, then "arrogant" California winemakers will be forced to listen and produce wines with lower ETOH. But it is also "arrogant" for you to say that people who like (and purchase) wines with higher ETOH like "bad" wines. From my perspective, if you like white zin with 3%RS then that is a good wine

W. Blake Gray said...

Andy: I try to balance my arrogance with humility. Don't always succeed, of course.

Many anti-alcohol campaigners are strident, and there's an understandable disgust about them from winemakers. But many, many winemakers (commercial, not necessarily home) want to blow off the issue by saying it only matters to a strident few.

I am NOT saying all wines with less alcohol are better than all wines with more alcohol. I drink every day, though, and all other factors being equal, I would rather drink wines with less alcohol, mainly because I love to drink but don't love being drunk.

What Larry Brooks and Jim Klein say about better winemaking leading to more alcohol in some cases, I get that. But there are plenty of less thoughtful and experienced winemakers ... ah, I've said all this before. And I'm late to a party. Where I'll be looking for bubbly and Riesling and, well, you know.

Adam Lee said...


It isn't just US winemakers that are supposed to list alcohol correctly on either side of the 14% line....but that listing is supposed to be correct for wines imported to the US as well.

I think this article showed that the CA winemakers were pretty good at it:

As far as you prefering lower alcohol wines because you like to drink but don't like to get much do you drink in an evening? You'd have to drink an awful lot, pretty quickly, for a 13% wine to have noticeably less of an effect than a 15% wine.

I loved the report, btw, thanks so much for pointing it out.

Adam Lee
Siduri Wines

W. Blake Gray said...

Adam: I like to start drinking about 6 p.m. and continue through maybe 10 p.m. -- that's my ideal evening. Well, that and having two baseball games on, a great meal, temperate weather and a bunch of friends. Oh, and world peace.