Friday, May 2, 2008

I Was Going To Post About Nutritional Information For Craft Beers

The problem is – it's not available. But what I really didn't know was that I was stepping into a decade long hornet's nest (mixed metaphors? Me? Never!).


This is a battle that has been on-going since 1993, when the Bureau of Alcohol, Tobacco and Firearms ("BATF") determined that "there was no significant consumer interest in having nutrition information on alcohol beverage labels." Then, in 2003, the issue was raised again when a brewery contacted the Alcohol and Tobacco Tax and Trade Bureau (TTB, the latest incarnation of what used to be the BATF) desiring to voluntarily put nutrition information on their labels. Then in December of that year the Center for Science in the Public Interest (CSPI) and others formally petitioned the TTB to change regulations to require nutritional information on alcohol labels. But, don't go thanking them just yet – their interests may not be entirely in your interest. The Beer Activist Blog notes that "CSPI has long espoused the dangers of alcohol and is what many beer supporters would refer to as a neo-prohibitionist organization. I like some of the work they do, including this awesome table of alcohol industry donations to members of congress, but their overall attitude toward alcohol is that it is ruining America." This petition requested that that beer labels include the following "Alcohol Facts":

  • Alcohol content
  • Amount of alcohol in fluid ounces
  • Number of calories
  • Ingredients
  • Number of servings per container
  • The current definition of "moderate drinking" for both men and women published by the US Department of Health and Human Services and Department of Agriculture

So, what do you think of these "Alcohol Facts"?

As a consumer, it doesn't seem to me that there is anything too objectionable. Well. Besides that "definition" of "moderate drinking" (currently defined as "one drink per day for women and two drinks per day for men).

Yet, various groups have various objections. For instance, the Beer Institute, the primary beer industry trade group, objects to publishing alcohol content. This group would prefer to keep this information from you because "The alcohol content in most beers is in a very narrow range, and consumers are generally aware of that fact." Like the Beer Activist, I call shenanigans on that one. Beers can range from less than 4% ABV to over 20% ABV, while maybe a 16% range is "narrow" to some, it seems like a pretty broad range to me.

Why does the beer industry promote this theory? Well, "The fact is that distilled spirits can contain up to ten times as much alcohol per ounce as a can or bottle of beer. A whiskey drinker can consume as much alcohol in one swallow as a beer drinker consumes in one beer. Most cocktails that Americans will recognize contain more alcohol than a bottle of beer; many contain twice as much. … The misguided equivalence campaign favors an industry that imports most of its product and contributes disproportionately fewer jobs and tax dollars to the U.S. economy when compared to the beer industry." In plain English? The general theory for taxing alcohol is for the industry to help pay for some of the damage that it causes (e.g., increased hospitalization, drunk driving, etc.) – and the beer industry's stance is that by taxing by volume rather than total alcohol output the beer industry is disproportionately taxed for its relative contribution to public drunkenness. For example, in twelve ounces, you can have 4% ABV of beer, or you can have about a third of a bottle of tequila at 40% ABV. What the left hand doesn't want you to remember is that your average drinker drinks more beer than liquor. Moreover, some of these Beer Institute members are talking out of both sides of their mouth, as Anheuser-Busch, MillerSAB, and InBev (the world's largest beer distributors) are aggressively adding to their liquor holdings as well.

This issue also affects what an appropriate serving size is. I would argue that an appropriate serving vessel for the Dogfish Head 120 IPA, a 21% ABV beer, is a champagne flute. In that case a single twelve ounce bottle is two servings, not one. To get the same alcohol content of one twelve ounce bottle of the 120 IPA, you could drink almost a whole six pack of Miller Lite (which one looks like the better deal now?) This all means that the "typical" nutrition information is effectively cut in half – making beer look almost downright healthy!

Then, in 2005, the TTB finally published a proposed rule, setting out the standards for putting labels on bottles of alcohol. There was significant disagreement among the commenters at this time about whether the labels should be mandatory or optional. The medical associations, CSPI and distillers all favored mandatory labeling. The brewing industry preferred that labeling remain optional; ostensibly because the labels would increase costs. There was also disagreement whether to list alcohol content by percent (as is traditional) or by actual volume (about .6 fluid ounces per 12 ounces of "regular" beer, 5 ounces of wine and 1.5 ounces of 80 proof liquor). Recognizing the wide range of alcohol content of beer, the US Bureau of Consumer Protection noted that "Beers in the marketplace range from approximately 3.3% to 17% ABV, thus delivering between .39 and 2 ounces of alcohol per [12 ounce] serving." Perhaps thinking that consumers may be interested to see exactly how little alcohol is in a 12 ounce "regular" beer (only .6 of an ounce? That's not much is it?), the CSPI withdrew their request to display alcohol content by actual volume. The breweries were also in favor of a percentage approach mainly because "a standard serving size is not consistent with the manner in which many alcohol beverages are actually consumed."

During the course of the 2005 process, the comments clearly indicated that "consumers are very interested in having information about the calorie and nutrient content of the alcohol beverage products they purchase. These consumers expressed the view that this information should be available on the product's label."

The latest proposal, issued in July of 2007, merely amends the 2005 proposal to solidify the requirements around the concerns raised to the 2005 proposal – comments for this 2007 proposal ended at the end of January 2008. This proposal would require a "Serving Facts" label that displays alcohol content by volume (disclosure of ounces of alcohol would be voluntary), the serving size (defined as the amount of the alcohol beverage customarily consumed as a single serving – for beer less than 10% ABV, 1 serving is 12 ounces, over 10% ABV, 1 serving is 5 ounces), calories in the serving, carbohydrates, fat, and protein. This proposal would have a delayed effective date to allow a three year implementation period so that small businesses could use up existing label stock and get new labels produced that conform with the requirements.

While the government deleted a comment from "J. Citizen", as a "J. Citizen" myself, maybe I can use my own forum to comment. Really, more information can't be a bad thing. Unfortunately, so many companies use these labels for misleading purposes (e.g., by using unrealistically small serving sizes, etc.). But, as Starbucks is finding in New York City where all chain restaurants are now required to display calorie, carb and fat information at the point of retail sale, more information doesn't necessarily equate to lower sales. People just want to be informed of what they are putting in their bodies. Moreover, the definition of serving size promulgated by the TTB seems eminently reasonable. While quite a few people will drink an entire bottle of, say, the Dogfish Head 120 IPA at 21% ABV by themselves, it doesn't seem absurd to recognize this as, at least, two servings. Interestingly, the craft beer industry suggests (pdf) even more specific serving sizes: 12 ounces for less than 10%, 6 ounces for 10-20%, and a mere 2 ounce serving size for beers over 20% ABV – a shot of 120 IPA anyone? While I applaud the effort, I think that this breakdown unduly "rewards" high alcohol beers (basically divides calorie and carb information in six!) and is not particularly realistic (to wit: the aforementioned champagne flute of 120 IPA).Where the government gets it wrong, I think, is not requiring full ingredient information. There is some suggestion that this is intentional so that people don't realize just how unprocessed some beers are as compared to others. But again, if we view this as a consumer protection regulation, don't you want to know that the beer you are drinking contains more corn than barley? Or more "lemon flavoring" (and the components of that "flavoring") than actual lemons. I recognize the "hassle" of re-tooling bottling lines, or more specifically the labeling portion of the bottling lines, to accommodate this regulation. However, the phase-in period seems reasonable and, really, how much of a hassle can it be? I don't see how it requires anything more than moving some text around on the label to accommodate the "Serving Facts" and printing the label. Perhaps the bigger hassle is actually learning how to calculate this information? Indeed the Brewers Association suggests that reporting tolerances be expanded so that small breweries without the labs of the bigger breweries don't "accidentally" violate the regulation.

Recognizing the inevitability of this regulation, the brewing industry also finally recommends that they be allowed to use as small of a label as possible, preferably a linear display over a full "serving facts" panel.

1 comment:

  1. I'd like to see one drink defined as 0.6 oz of alcohol, all beer servings fixed at 12 oz and labels that state the number of drinks per serving.

    Since defining moderate drinking for everyone as x servings of alcohol per y lbs of body weight would make too much sense, maybe the gender bias could be reduced by making a serving of alcohol 0.6 oz for men and 1.2 oz for women.


Note: Only a member of this blog may post a comment.