Felix Salmon

Charts of the day, wine-heat edition

By Felix Salmon
February 8, 2012

Last year, I blogged a paper about the way in which wineries lie about the alcohol content of their wines. Now, the same authors have a new paper out, trying to get to the bottom of exactly why wine is getting hotter.

One thing I like about this paper is that it doesn’t look directly at wine-alcohol levels, but moves back a step to the sugar content of the grapes going into the wine. If you turn high-sugar grapes into wine, one of two things has to happen: either you get sweet wine, or you get high-alcohol wine. Since taste in wine is getting dryer rather than sweeter over time, higher-sugar grapes mean hotter wine. And those grapes are definitely getting sweeter, especially when it comes to white wine. Here are the charts for California:


Is there a global-warming thing going on here? After all, warmer temperatures mean sweeter fruit. But, no. Here are the temperatures of California’s wine-growing regions for the years since 1990 that sweetness has been rising quite dramatically:


But we kinda knew this already. The most interesting thing in the paper, is not that hotter wine is unrelated to global warming. Instead, it’s that hotter wine is quite strongly related to price:

Sugar content of grapes at harvest was relatively high for red varieties and premium varieties, and for grapes from ultra-premium and premium regions. The same categories tended to show evidence of faster growth rates in sugar content as well… In all of the models, the analysis shows a higher propensity for growth in sugar content for premium varieties, compared with non-premium varieties… This feature and the patterns of the level of sugar content among regions and varieties could be consistent with a “Parker effect” where higher sugar content is an unintended consequence of wineries responding to market demand and seeking riper flavored, more-intense wines through longer hang times…

We found that the region with the lowest price of wine grapes (under $500 per ton) had significantly lower average degrees Brix at crush compared with all other regions… It may be profitable, in producing lower-priced wines, to opt for a higher yield of wine per ton of grapes in exchange for lower Brix.

Simplifying, you can think of vineyards in one of two ways. Either they’re a source of grapes which get sold by the ton, in which case you want to maximize the yield. That, in turn, means lower sugar content. Alternatively, they can be a source of carefully-cultivated grapes which get turned into premium wine selling for $20 per bottle and up. In that case, the quality of the grapes starts trumping their quantity. And it’s pretty clear that what winemakers want, if they’re going to sell expensive wine, is grapes with a lot of sugar. That’s their expressed preference, anyway.

All of this is consistent with what I wrote last year — that wine drinkers say that they want lower-alcohol wines, and will even go so far as to prefer to buy wines with lower alcohol numbers on the label. But when they actually taste the stuff, in general the higher the alcohol the happier they are. Especially when the wine is expensive.

7 comments so far | RSS Comments RSS

Felix, nice post. Just out of curiosity, I wonder what the precipitation patterns look like over this period, as well. Not that I doubt the explanatory power of your argument, just interested in potential confounders.

Posted by Josephus3 | Report as abusive

Higher alcohol might be correlated with happier customers, but is it causative?

Felix writes: “when [wine drinkers] actually taste the stuff, in general the higher the alcohol the happier they are.”

But another alternative is in the quote Felix includes: “could be consistent with a “Parker effect” where higher sugar content is an unintended consequence of wineries responding to market demand and seeking riper flavored, more-intense wines through longer hang times…”

If wine drinkers prefer “riper flavored, more-intense wines”, and growers grow that fruit, and wine-makers make that wine, couldn’t high alcohol be a by-product in the search for fruit taste, rather than the goal? Perhaps customers enjoy the flavor, but observers conflate that preference with one for high alcohol content?

Correlation among alcohol and preference, not causation.

Try a blind taste test of a specific grape from a specific region: full-flavored, moderate-alcohol wines vs. full-flavored, high-alcohol wines.

Posted by SteveHamlin | Report as abusive

Averages will go up if the top values go up, if the bottom values go up, or if all values go up. If you look at the graphs, the trend for wine has stabilized around 23 brix, and 22.5 for white wine. If you are trying to make a high-quality red, you will want to reach 24 brix typically. Whites might be harvested a bit lower, and grapes for sparkling lower yet. But 23 brix is probably where the ripeness should be. The increase in the trend line is partly, yes, a fashion for more concentrated flavor in wine that comes along with greater ripeness and greater alcohol, but the trend also reflects a lower tolerance for poor quality grapes delivered to the crush pad. Grapes were grown in the 80′s very often in “California Sprawl” fashion, with lots of irrigation. Vertical trellises and deficit irrigation are much more common today, with smaller, less vigorous canopies, more light and air on the grapes, resulting in grapes that have more concentrated flavor components, and yes, more sugar and therefore more alcohol. In other words, the average has gone up because the bar at the bottom has come up, for which we should be thankful. One other thing to note is that the brix generally goes up a bit a couple points in the tank and then the alcohol is usually about 1/2 the brix achieved in the tank. You are not going to make a 14+ alcohol wine from grapes that are 23 brix. Much of the talk of wine being too alcoholic really concerns wines that are substantially more than 14% alcohol and are in the “ultra-premium” wine category. The great sea of inexpensive wine out there comes nowhere close.

Posted by kenms | Report as abusive

i’m only a dog, but….

Can we please end the distracting and incorrect hyphenation of adverb-adjective combinations? A hyphen is properly used when several nouns or a combination of nouns and adjectives is used as an adjective; for example “adverb-adjective combinations”.

Incorrect uses from the article: “carefully-cultivated grapes” (Felix) and “more-intense wines” (article quoted). These both combine an adverb and an adjective and should not be hyphenated.

Posted by samadamsthedog | Report as abusive

We’ve noticed the higher sugar levels and the related higher alcohol levels and we’ve cut back on our wine drinking. It’s hard to drink more than a few sips of anything from Napa Valley these days, and we used to lov Napa Valley wines. If we want something sweet and alcoholic, we’ll mix up some Test Pilots or Mai Tais.

Our impression is that Americans really do like sweet, high alcohol wines. They don’t like dry wines where you can taste the grapes, tannins, spice notes and the like. They want a fruit bomb, so the market has been delivering fruit bombs. A lot of the brix is when you harvest your grapes, not the varietal itself, so vintners harvest later.

Robert Parker is just in the middle of the storm. American tastes have moved to sweet, and I’ll bet a lot of new wine drinkers around the world appreciate sweet wines as well. They are much more accessible if you’ve never had wine before. Despite this, we do manage to find a number of old fashioned wines that are relatively dry, but it takes some doing. We are fighting to popular trend.

Manischewitz was just ahead of its time.

Posted by spiffy76 | Report as abusive

In Bordeaux, it’s the Michel Rolland Effect, which is the transmission mechanism for the Parker Effect. Long hang time, ripe grapes are his recommendation to every client. Quite boring….

Posted by maynardGkeynes | Report as abusive

I agree with the above comment. The alcohol level in mass market wines in general is going up without any additional benefits. With the best of Bordeaux aiming at 12% traditionally with wines delivering complexity, character and finish at 12% why do we need 14-15% ? this is fortified wine level

Posted by WinedayUk | Report as abusive

Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/