Wine labels and alcohol content
What is the precision?
Wine labeling, regulated by legislation, ensures that the consumer is provided with facts about alcohol percentage and the number of ‘standard drinks’. Or does it? What do those numbers mean?
Numbers have a habit of looking scientific, especially if they have a decimal point. So, a red wine label stating 14.5% per volume (Australian wines are generally high-alcohol) suggests that the red wine contains 14.5% alcohol by volume (ABV) — right?
Not exactly. It means that the alcohol content can be as high as 16% and as low as 13%. Labelling regulations allow a 1.5% variation from the percentage given on the label.
The label could easily give a range rather than implying the precision of a number. Given this range, it is misleading (and unscientific) to give a figure of 14.5, which suggests accuracy to at least one decimal place. It also doesn’t mean that on average, the percent is 14.5. On average, the percent is still allowed to be 16. The 1.5% is not a measurement error — it is an allowed reporting range.
The reason for this 1.5% leeway is not obvious. Certainly, it is scientifically possible to measure to nearly any level of accuracy (including 0.5%). The only suggestion I have seen is that vintages vary, but everything coming out of a vat and going into a bottle should be fairly uniform I would think. Or, at least, the consumer could be informed.
The Australian regulations for fortified wine (sherry, port etc) require a more stringent 0.5% margin. So, clearly, greater accuracy is possible.
The 1.5% leeway provides a potential for manipulation. Higher alcohol wines may suit a vintner’s purpose (e.g. enhanced mouthfeel), but are less culturally-correct in these times of increased alcohol awareness. So, a wine could be produced at 16%, but legally marketed as 14.5%. I would not want to be so cynical as to suggest such a thing occurs.
What if the vintner wants to decrease alcohol level for a given vintage? The obvious thing to do is to add some water to the wine. Given that wine is 70–85 % water, this does not seem unreasonable — water is the most naturally abundant component of wine. But, no. In Australia, adding water to wine for the purpose of reducing alcohol level is illegal. A small amount (no more that 7%) can be used to dissolve additives that a vintner might choose to add to the wine. If a substance is not on the list of legal additives, then it cannot be added to wine.
What about standard drinks? A standard drink contains 10g alcohol. Note the switch from percentage by volume (ABV) to actual weight of ethanol. Is that useful? A standard drink does not mean a drink of a standard size. The concept has never really taken off.
Regulation stipulates that the number of standard drinks be given to one decimal place. You can perhaps see where this is leading — one decimal place is much more accurate than the 1.5% range for ABV. An accuracy of one decimal place for standard drinks would require ABV to be accurate to better than 0.2%, not the allowed 1.5%.
But, remarkably, standard drinks are calculated from the ABV. This means the decimal place in standard drinks is misleading too, because the leeway in ABV is much greater than the precision implied by one decimal place. How this helps a consumer is not obvious. It brings me back to my opening comment — a decimal place (for both ABV and standard drinks) is probably there to imply precision, not to provide it.
The number of standard drinks is calculated from the volume of the bottle (in litres) times ABV times 0.789 (the specific gravity of ethanol at 25C). Thus an ABV of 14.5% would see a wine label read 8.6 standard drinks, but it could contain anything in the range 7.7 to 9.5.
Head spinning? Maybe pour yourself a glass of wine, ignoring the label. This is my recommendation — the numbers really don’t mean anything. Would you chose a 13.5% wine in preference to a 14.5% one because the ABV is lower? No, because legally they could both be 15% (or 13%). The 14.5% wine could even have a lower (real) percentage than the 13.5% one. Choose a bottle with 7.7 standard drinks over one with 8.7? — no, they could both contain 9.5.
If ignoring the label is too much, consider taking some control. Wearing your cynical hat, and assuming that an Australian red labeled 14.5% could probably be 16%, and that you would prefer to drink something with your meal at more European levels of 12%, then for a 750ml bottle of wine, add 250ml of water. ‘Breathe’ in a blender. Done. You have reduced the cost of the wine (by volume) by 25% and now have 1L of wine rather than 750ml, you have a perfectly agreeable wine to accompany food and you looked after your health by reducing ABV — label notwithstanding.
You can choose the dilution yourself. The general formula to reduce ABV is: Take a stab at what the ABV of the wine might be (given the number on the label) and subtract the desired ABV from that, multiply the result by the bottle size and divide by the desired ABV. This gives the volume of water to add.
Adding water to alcohol can actually improve flavour for higher-alcohol beverages. As Harold McGee explains, aroma molecules cling to alcohol. By diluting ABV, these molecules are more likely to be released and travel retro-nasally to be perceived as flavour. There are tradeoffs of course, and the ideal is to produce the wine with a lower ABV to start with.
To quote Harold:
“High-alcohol wines, those that exceed about 14 percent alcohol, are often described as “hot” and unbalanced. Alcohol’s irritating effects account for the heat. And flavor chemists have found that high alcohol levels accentuate a wine’s bitterness, reduce its apparent acidity and diminish the release of most aroma molecules. Alcohol particularly holds down fruity and floral aromas, so the aroma that’s left is mainly woody, herbaceous and vegetal.”