Thursday, May 16, 2013

Note to wineries: Label alcohol percentage accurately

Wine labels in the US are allowed to be wrong about the alcohol percentage by up to 1.5%. There used to be a solid legal reason for this. But the government has adjusted the law, and it's time for the wine industry to adjust as well

Currently wines under 14% are allowed to be mislabeled by up to 1.5%, while wines over 14% alcohol can be mislabeled by up to 1%. This means a wine labeled at 12.5% alcohol can be anywhere between 11% and 14%, while a wine labeled at 14.9% alcohol can be anywhere between 14% and 15.9%.

This is why so many wines are labeled at 12.5% and 14.9%. It's a hedge, and until recently it was a reasonable one. The federal agency responsible for approving wine labels, the TTB, previously made it difficult for wineries to make any changes to labels without going through the time-consuming, unpredictable approval process again.

But now the TTB has had three years of staff reductions and has simplified its approval process. Wineries can change a lot of things on the label without new approval being required -- including alcohol percentage. Wine label alcohol percentages can go over or under the 14% dividing line without a new approval.

There's just no excuse anymore to be so inaccurate.

Sure, some consumers don't care about alcohol percentage. But many do. That's the reason federal law requires the alcohol percentage for wines over 14% to be on the label in the first place.

Germany has a tolerance of just 0.5% on its labels. It's the 4th largest wine market in the world, so importers take the trouble to get the alcohol percentage right.

The US is the world's largest wine market. Don't our consumers deserve the same accuracy as German consumers?

Wineries: Nobody is telling you to purposely make your wines higher or lower in alcohol; both styles have fans. But there's a huge difference between a wine that's 14% alcohol and one at 15.9%, and it's not right that we don't know which is which.

Just be honest with us. That's all we ask.

Follow me on Twitter: @wblakegray and like The Gray Report on Facebook.


Anonymous said...


Did you consider State Labeling/Licensing requirements?

Honestly, it hasn't been a Federal (TTB) concern for quite some time. The issue is more a state concern. There are states which require that each change in vintage needs a new approval. Other states that require a new approval with each new TTB approval. Other states require a new registration with each price change. Other states say that a material change in the label (interpreted by many as including alcohol levels) require a new label approval. And on and on....Each of these registrations costs us money. Last year I spent over $37,500 in compliance fees (both state and federal) and that doesn't include the fees that we pay the compliance company for doing the work.

Just one example, the state of Alabama has two label of which has approved the Siduri label and the other has not. So we have some of our wines that we can sell in the state, but others we can not.

Germany (your example) has the same laws in all of its states regarding alcohol labeling. That is not the case here in the United States.

I am not against stronger alcohol labeling laws (along the Germany range) but would far prefer it if a chance in this regard could remove any cost increases to wineries on the state level as well.

Adam Lee
Siduri Wines

Ben said...

Really dude? This bothers you? Gimme a break. Blog about something of significance. Unless you are hell bent on getting trashed, that little bit of a diffence isn't going to matter one bit. I never even look at the alcohol percentage because I have no intention of getting drunk so it doesn't matter. And who cares what Germany or any other country does. This is the USA. If you want to hold a micrometer up to everything go live over there where you can get a good night sleep knowing that your gluvine is exactly 13.7789% alcohol by volume.

W. Blake Gray said...

Ben: If you have no intention of getting drunk, it does matter.

Ah yea, the "who cares what any other country does" argument, followed by "if you disagree with any part of my diatribe, leave the country permanently." Because the country is yours to rule.

Unknown said...

Yeah, Blake. If you love Germany so much, why don't you marry it......dude?

W. Blake Gray said...

Kent: That marriage wouldn't be legal in California. Yet.

Bobby F said...

Another aspect to the labeling, of course, involves tax. In states like Georgia, there is a higher tax for wines above 14.1%, which can significantly change the price of a less expensive wine.

Whealdeal said...


Dumb question, but what if your wine sits at exactly 14.0%. Are you give 1.5% or 1.0% flexibility?

Anonymous said...

I read this post and then right after read Tom Wark's post on DUI changes. Seems that these two topics are very related. The difference between a 15% abv and 12.5 is massive if your out to dinner and driving home as Tom showed in his post.

Man About Wine said...

Wheal, the Fed rule is up to 14%, you are in the lower excise tax group and you have 1.5% leeway. Over 14%, the higher tax rate, and 1% leeway. And realize that precise measuring is not easy, changes with time, and with evaporation related to temperature, humidity, and atmos. pressure. So, in practice, wineries at 14% either choose to pay the higher tax rate and avoid the potential of a federal violation, or they will water down the wine by 1 or 2 tenths, to get down to 13.8%. which is why you see a lot of mass market wines labeled 13.9 or 14.

I am OK with the current law leeway. I wish the SIZE and CONTRAST of the label were forced up to make it harder to hide the %. Ridge is the positive role model for me, they print the grape blend and alc. % in easy to read block print.

Unknown said...


To be more precise, wines containing 14% or less alcohol have a +/- 1.5% margin of error. Wines containing more than 14% have a +/- 1.0% margin. A little known fact is that wines containing 14% or less alcohol are not even required to list the alcohol, as long as they display “table wine” or “light wine.” I’ve seen examples of this on some labels. Another little-known, permitted option is to display a range – max. 3% range for those 14% or less and 2% for those over 14%.

Larry Brooks said...

While I'm not sure I agree with Ben's tone I think the gist of his comment is correct. I also think this entire alcohol accuracy thing is a bit of a stalking horse for the true subject which is the flavor of excessive ripeness. I'm pretty sure Dan Berger was one of the first to really start hammering wines for high alcohol, but in reality I think what he and many tasters object to is the flavor of high ripeness. Granted the two often go hand in hand, but I've made wines with very ripe flavors and moderate alcohols such as the '04 Pinot Noirs and I've made several white wines over the years with alcohols around 15% that tasted fresh and balanced. Maybe what we need is a ripeness index. When anyone asks how much alcohol is in my wines my stock response is "sufficient"

Jonas Landau, everydaywineguy said...

@Ben - Don't know why this post would bother you so much. There's a lot of information about the wine in the alcohol level. It matters a lot to me as it tells me how lean, ripe or over ripe the wine may be.

Stillman Brown said...

The state requirements really are a pain, another legacy of the improper repeal of Prohibition, besides letting the mob go legit. I make wines ranging from 11.9 to 18.0 - if the customer is going to drink more than a glass, they ought to take a look.
When I'm a consumer myself, I always want to know. It might be more important than the price or even the 100-point score.

Unknown said...

Blake, Alcohol levels in the cellar are a bit dynamic. This gives rise to two issues:
#1 For a small winery label printing is expensive and gets much more expensive if you run small batches. Per label cost for 70 cases costs more than double the per label cost for 600 cases. I have a simple label but unless I want to raise my prices by more than a dollar per bottle I need to print all of my labels for each vintage in one run even though some of the wines will remain in barrel for more than six months after printing.
#2. In my case, labels must be applied at the bottling line. Imagine how much more the wine would cost if we ran alcohol numbers at the bottling line, then bottled shiners, and then waited a month for the label to print (no bonded facility on site so movement charges would crush us), then needed to run a labeling line. Think about the logistics of it all. When we turn in our labels - not just to TTB but to the printer - the final blends might not even be assembled (a few gallons from one lot blended into another can make a world of difference in the wine).
We do our best to get the numbers as close as we can, but it isn't practicable to be precise. Does it really matter so much that you would be willing to make every consumer pay more for their wines and have fewer wines available?

W. Blake Gray said...

LW: I've heard this argument before, but how much variance are we talking about? Will the wine really vary more than 0.5% from batch to batch?

Re printing labels in small batches: You have to print a new label with the vintage for each wine anyway. So unless the alcohol label varies by more than 0.5% in the batch itself, I don't see how this applies.

Look, I don't want to see life made harder for small wineries, but I don't understand how in this era small wineries can't calculate the alcohol within 0.5% in time to order labels printed. Batch variation would seem to be a much bigger issue for giant industrial wineries, and they get print discounts.

Anonymous said...


I do think 0.5% margin either way is fair.

I do also think it would result in higher costs for small wineries that distribute their wines, given what I mentioned about state licensing, etc.

You may not be trying to make life harder for small wineries, but you are advocating making it more expensive for them.

Adam Lee
Siduri Wines

kschlach said...

Blake, I believe LW is referring to the fact that many small wineries bottle the same wine from a vintage in different batches (sometimes more than a year apart). This could create different alc levels. Nevertheless, I side with you on this argument...

Coop said...

Man, sometimes its really hard to nail that alcohol right on the head. I work at a very large facility and sometimes the labels may be printed in advance based off a bench blend or calculated from components of existing lots.

That's not accounting for a large blend that gets shorted and has to be brought to volume again, or a cellar worker has a accident and something is watered back! Yikes!

Plus the technology to be precise when measuring ALC is costly. Using a classic/electric ebulliometer will give you inaccurate results only a density meter or NIR tech will get you close to absolute, and that's about 20,000 to 50,000

Of course you can send stuff out to a TTB approved lab for testing for a cheaper cost, but if you plan on expanding your blend well..

Sorry for the ramble! In short you can be accurate with you ALC on labeling but it's really a bitch.