Many a wine consumer has tasted a wine and then looked at the alcohol level and thought, “That doesn’t seem right.” There are many reasons why that suspicion might be correct.
Wines fourteen percent alcohol by volume (ABV) and under are allowed a 1.5% margin of error – provided the alcohol content does not exceed 14%. Wines over 14% are allowed a margin of error of 1%. In other words, a wine listed at 15% alcohol may be as high as 16% without running afoul of the feds. For this reason alone, there is a good chance that the percentage listed on the label is not entirely accurate. There are other reasons.
Wineries sometimes have their labels made before they have their final ABV numbers. There’s no problem with this – as long as the final alcohol level is within the allowable margin of error. However, if the level is beyond the margin of error, sometimes wineries don’t pay to make new labels. I first noticed this happening when I saw discrepancies between technical sheet and label information.
Some wineries seem to just put the same number on the label every time on every wine (14.5%!) year after year. Presumably, what is on the label is close enough to the actual ABV level so who cares? Doing otherwise means getting new labels made. I recently tried a lineup of wines from a winery where every wine had the exact same alcohol level listed. What are the chances?
Taxation is also an important consideration. The Alcohol and Tobacco Tax and Trade Bureau (TTB) taxes wine at 14% ABV and under at $1.07 per gallon; wine over 14% is taxed $1.57 per gallon. Wineries therefore have a financial incentive to label their wines below 14% ABV. Do some try to take advantage of this? No idea, but I wouldn’t be surprised.
I also have a suspicion that some wineries like to err on the side of listing lower ABV levels than what is actually in the bottle. I have tried numerous wines where I thought the percent alcohol was considerably higher than what was listed. Though that is only anecdotal, I know several winemakers who have sampled a bottle of wine, seen the alcohol level listed, and subsequently tested the wine and found that the ABV was considerably higher than what was listed (and not necessarily within the margin of error).
Are some wineries consciously listing a lower alcohol percentage than what is actually in the bottle? Alcohol levels in Washington State and elsewhere have increased dramatically over the last ten years. Are wineries willing to own up to these ABV levels or do they fear a consumer or critical backlash? I don’t know, but, anecdotally, there seems to be many wines that have higher alcohol levels than what they have listed.
How widespread are these issues? I have no idea frankly. I expect that larger wineries are more exact with their ABV numbers than smaller wineries both because they have better equipment and because they have more to lose if they are in error. Presumably these wineries are also more likely to have their wines checked by the TTB, although I have no idea how frequently these types of checks are done.
Personally, I am of the opinion that the alcohol level printed on a wine label can be useful information, whether saying something about the potential crispness of a riesling or the opulence of a cabernet. I would love to see the labels accurately portray what is in the bottle. But, for now, look at those labels with some skepticism. Many times the numbers listed are not exact.
Sean,
Your article mentioned the tax rates for the Feds, the same kind of tax applies to the state as well. Wineries pay to WA State $0.8676 per gallon of under 14% and $1.717 per gallon of over 14%. The label tolerance ensures that if I have a 13.1% tested wine and I am confident that I did not make a mistake, I can slap it on the label, but it if is audited and tested over 14, then the wine is in a new tax bracket and any taxes paid at the under 14 rate will need to be adjusted with penalties.
The safety net is to put 12.5% on the label. That is what I am doing and putting the exact measurement results on our website. Also, wine under 14% can also be labeled as Table Wine, so there is no indicator at all there. The tolerance is a bit of a buffer to address evaporation and such so, as you mentioned, a winery can make up label in advance of bottling.
Over 14% you have to be a bit more accurate as you mentioned, 1%, but if the wine is 14.1% tested you cannot go down a tax bracket. So for me, I error on the side of over 14% be a close to the numbers I got in my testing. I use three methods.. Starting Gravity – Ending Gravity, ..Starting Gravity * a conversion constant, and finally Dr. Honeyman methodology (most accurate).
I don’t think wineries are trying to dupe anyone, but I know that I rather be paying less taxes. Side note, I really don’t care about the debate on what if the wine was 15.1% and I put 14.1%..it is all about the tax for me.
Micah, great stuff. Very helpful. Thanks for the thoughts.
Sean,
I do believe that many wineries 'fudge down' as much as the government allows them – if it's 14.1 on the label, more often than not the actual alcohols are north of 14.5 . . .
One other huge point to make – I wonder if the government monitors alcohol levels as closely as consumers believe. I do believe they look closely at tax class clarifications – for instance, if you say a wine is under 14%, you better have it that way (or the government will be losing out revenue). But if it's north of 14%, my gut feeling is that the government may not be enforcing the 1% rule all that often . . . and probably even less so on imports.
Cheers!
Larry Schaffer
I am well aware of one winery that got caught putting an abv of 14.1 on the label and the Feds decided that the wine was <14%. Said winery had to pull the wine off shelves until a "use-up" permission was granted (about a month of no shelf space.) The winery, as well as a couple of certified labs, said the wine was north of 14% but the Feds, in their ancient method of using distillation to check alcohol, decided the wine came in at 13.95% v/v. No matter what TTB certified labs said, winery had to go by results from Walnut Creek lab.
Interesting . . .and I wonder what the 'true story' was. I can't imagine the Feds trying to fight that hard, especially when they were, in essence, going to lose money on the deal (less tax revenue under 14 than over). If the winery used a 'certified lab', I think the government would have a hard time fighting the results . . .
Anything else to add? Had this winery been 'busted' before or under scrutiny for anything else?!?!?
Cheers!
Larry, my thinking was similar. The government is most likely more concerned about the tax bracket than the alcohol level. If so, why make a stipulation that it has to be within 1%? I don't know if this is true but it would seem to make sense.
Anon, fascinating story. Thanks very much for sharing it. I've been asking around to folks trying to find and example of this. I'll do some checking into how the feds check alcohol levels too. VERY bizarre that they would make a big deal with a wine on the low side, especially that close to 14%.
I know people who shop wine based upon ABV and won't buy a red (other than Pinot Noir) that is south of 14%. I don't know why they came to that decision point but it is a benchmark for the consumer with certain red varietals and blends.
Terry, I definitely know consumers who use these numbers to inform their purchasing decisions as well – both on the low side and the high side. Part of why I think it's interesting that the numbers are not necessarily right!
Being in the label printing industry, I see this daily. Sure, there's a cost involved in reprinting the label, but there's a cost involved in changing the alcohol from year to yeat as well. Sometimes the alcohol might be printed in a different color than anything else that might change, so correcting the alcohol each time is what can be considered an unnecessary expense, especially when you can set it as 13.5 and forget about it.
Imagine, too, that the alcohol can actually CHANGE between testing, ordering labels and bottling. I'm afraid you will have to live with an estimation unless you have your own lab to test it from the bottle.
Anon, thanks for the insights. Much appreciated.
There are more consumers than wineries, so the perfect solution would be to have the consumers petition , picket, etc. the government to change the alcohol rules to a sliding scale, instead of
a under/over 14 scale, with the whole range between 12.5 and 14.9 a danger zone of government penalties (see 13.95% example above). If enough people want the change, it will happen, or it hasn't happened because not enough people care.
Intentionally labeling a wine as <14% when it is truly above 14% is considered tax evasion. I doubt many wineries would go there. Federal law (read the CFR) requires you to own an ebulliometer. You can get samples to test the accuracy of your ebulliometer from the feds or any lab.
And that is precisely the problem. An ebulliometer is required by law to measure alcohol. However, these devices, as with any other measurement instrument, carry an inherent accuracy and precision that, in this case, is not particularly good. The reasons why this instrument is such a terrible measurement device are many and beyond the purpose of this comment. I've often seen a repeatability (precision) range of more 1.5% abv. I suspect this is the reason the TTB allows such a wide margin. The problem arises when the TTB also requires the instrument to discern a perfectly defined tax bracket threshold. This is off course impossible and fraught with peril.
An interesting post on this subject with discussion of testing results from Dr. Vino from a while back. http://bit.ly/hQqk2c
As backwards as it seems, the Fed is absolutely concerned with inaccurate labels – regardless of the direction of the error. The TTB routinely pulls wine from the shelves and test ABV at their own lab. It does not matter what documentation a winery has of ABV at the time of bottling, if the TTB lab generates numbers that put the wine in a different tax class than what is stated on the bottle, the winery is in volation. (And they DO follow up. And they DON'T give refunds…only slaps on the wrist.)
The biggest problem I see is that the 1% tolerance level only applies when the wine is >1% away from the tax class break. The magic number is 14.04%, meaning 14.05% ABV is in the over 14% tax class and 14.03% ABV is in the below 14% tax class. Therefore, if "Winery A" puts 14.00% ABV on the label, and the TTB lab reports the wine at 14.05% ABV, "Winery A" is out of compliance even though their stated alcohol is 99.6% accurate. Another winery, let's say "Winery B", puts 14.05% on the label and the TTB lab reports the alc at 15.00% ABV. Winery B is still in compliance with only 93.6% accuracy. Winery B gets by with a far less accurate number simply because the acohol concentration is farther away from the TTB's arbitrary tax break threshold. The 1% tolerance level is a misnomer. The tolerance level is anywhere between 0.01% and 1% and depends soley on what the TTB lab in Washington DC measures the alcohol to be.
Anon 3/14, very interesting stuff. Thanks for sharing it. I was wondering how the system worked when one approaches the magic 14% number. That is some VERY strange stuff that only the government could make up. I wonder what the accuracy and reliability is of the small winery ebulliometer? It would seem like airing on the high side would make 'sense' when near the threshold. Why am I not surprised that the government doesn't give refunds?
Sean-
Great post about this highly "popular" topic these days when it comes to wines and their actual vs. label alcohols.
As a compliance expert I've definitely seen the alcohol "dance" done come label design and approval time at many wineries. They do often shy away from putting a higher alcohol on their label (when in the higher tax class) and use up that 1.0% tolerance range to fudge it down. Is your take that consumers are really paying that close attention to label alcohols for their buying decisions?
Ann
Ann, I believe that a certain segment of consumers – wine geeks – pay a good deal of attention to the labeled alcohol level even if they know it has a fudge factor. I even know people who use it to inform their purchasing decisions. Especially since so many wineries in Washington are small production shops, many of these people are the types of consumers that wineries are targeting, or would like to be.
While the laws do not require wineries to be more accurate than is stipulated, I would personally like to see the information presented be as accurate as is sensible and possible from a consumer perspective.
An interesting recent article on this topic from the Guardian looking at 129,000 wines.
http://www.guardian.co.uk/lifeandstyle/2011/jul/05/wine-alcohol-strength-understated