Regulatory compliance is not something that the alcohol industry is good at

There has been some attention recently on what information should be mandated for the labels of alcohol containers (New wine label rules are making alcohol content more accurate ; Portman Group updates alcohol labelling guidelines). Personally, I think that alcohol can reasonably be treated as “food” (I put it in my body, with the intention of metabolizing it), and so it should contain all of the same information as other food, with the same levels of accuracy. How can anyone argue for an exception?

Anyway, this leads me to ask whether the alcohol industry has, in the past, been trustworthy to put accurate information on their labels, even given the very little they have been asked to put there. Interestingly, there are official data that can tell us, at least from the USA.

The US Alcohol and Tobacco Tax and Trade Bureau (part of the Department of the Treasury) has, at least in the past, run a Market Compliance survey, which they describe this way:

Established in 2008, TTB’s Alcohol Beverage Sampling Program (ABSP) is a random survey of products in the marketplace, where we:

• Verify that the labels on alcohol beverages contain adequate descriptive information,
• Confirm that the labels are not likely to mislead consumers, and
• Determine where compliance issues exist.

Each year we purchase products from the marketplace and bring them to our offices for label assessments, where we first evaluate them for compliance with our labeling regulations. Following the label assessments, we send the products to our laboratories for a series of chemical analyses, to assess whether the products themselves comply with the information displayed on the product labels.

The readily available data cover the years 2011—2016, listed separately for: distilled spirits, malt beverages, and wines. Over these six years, the TTB reports listed a total of 25 different non-compliance issues, involving: the label information, the bottle contents, and the actual label / container. I have listed these issues at the bottom on this post, so that you can see what is officially expected from the alcohol industry. The average number of distilled spirits examined per year was 230, the number of malt beverages was 182, and the number of wines was 138.

The results are shown in the graph. Each point represents one alcohol type, for each year sampled (horizontally), along with the percentage of the bottles that were deemed to be not compliant (vertically). Note that any given bottle could have more than one non-compliance issue. There were an average of 1.16 issues per non-compliant bottle for spirits, 1.27 for malt beverages, and 1.19 for wine bottles.

This is not too good, is it? In fact, it is terrible.

Things start off at their “best”, in 2011, with less than one-quarter of the bottles deemed to have at least one compliance issue; but how can this be the best that the alcohol industry in the USA can do? Moreover, it seems to get worse through time, with spirits reaching a limit of two-fifths (ie. going from 25% to 40%), although wine manages to stay at one-quarter. The malt beverages start off similar to wine (20%) but end up being like spirits (40%). Note that the malt beverages also have the highest number of issues per non-compliant bottle.

The most common issue for the distilled spirits was alcohol content that was greater than that stated on the label (55.3% of the 521 issues). The most common issues for the malt beverages were also alcohol content that was greater than stated on the label (27.8% of the 400 issues), plus non-mandatory information that differed from the COLA (ie. not an allowable label revision; 20.8% of issues). For wines, the most common issues were this same one about revised non-mandatory information (32.1% of the 209 issues), plus an alcohol content that put the contents into a different tax class than that stated on the label (22% of issues). [Note: this tax issue applies only to wines, not to spirits or malt beverages.]

Conclusion

So, the US wine industry can sit there knowing that they are doing better than their compatriots, in terms of not misleading the public with regard to the presentation of their alcoholic product. However, they cannot be too smug about this, can they? After all, they apparently succeed only three-quarters of the time. A 75% success rate may have been acceptable (or even good) when we were back at school, but it can hardly be acceptable in one’s profession. Prior to drinking, the label is all that the customer has, in terms of information. Wine commentators tend to take things at face value, so they cannot help us in this regard, either,

But, boy do those brewers need to get their act together — the spirits industry is not one for them to start emulating, in this case.

[Aside: Part of the problem here is that regulatory compliance is self-monitored — the people doing the work are the same people who are checking for compliance. This does not happen where human safety is concerned — for example, structural engineering requires that an independent group check for design compliance, not the group who created the design. This is supposed to stop bridges collapsing.]

---

Information (9)

• No Certificate of Label Approval (COLA)
• Mandatory information differs from the approved COLA
• Non-mandatory information differs from the COLA (labels that did not match their approved COLA due to changes that were not allowable revisions)
• Mandatory information is missing
• Mandatory information is incorrect
• Mandatory information is on wrong label
• Government Health Warning Statement has errors

Contents (10)

• Class, type, or statement of composition is incorrect
• Statement of average analysis or serving facts statement is missing or is incorrect
• Tax class (alcohol content that places the product in a different tax class than indicated by the label)
• Alcohol content: over (alcohol content that did not match the label and was outside regulatory tolerances) **
  
• Alcohol content: under (ditto) **
  
• Fill: over
• Fill: under
• Prohibited practices
• Adulterated
• Pesticides
• Sulfites
• Formula issue

Label / container (6)

• Type size / legibility
• Label not firmly affixed
• Label covered by another label (label is pasted over)
• Permit address issue
• Organic certification issue
• Distinctive liquor bottle

** The TTB notes state:

1. Distilled spirits generally allow for a loss of 0.15% alcohol by volume; however, no tolerance is allowed for an increase in alcohol by volume.
2. Malt beverages generally allow for a tolerance of 0.3% alcohol by volume, either above or below the alcohol content stated on the label.
3. Wines are allowed a tolerance of 1% alcohol by volume, either above or below the stated label alcohol content, for wine containing more than 14% alcohol by volume. Wines falling between 7% and 14% alcohol by volume are allowed a tolerance of 1.5% alcohol by volume. If at any point a wine crosses into a different tax class, the tolerances no longer apply — for example, if a wine is labeled as 13.5%, but is found to actually be 14.2%, we would consider this a violation.