Alcoholic strength on the label: legal tolerances and measured values
EU tolerances between declared and measured strength for wine and beer, data on 91,432 wines tested in Ontario and 218 UK beers, and measurement on the distillate.
The alcoholic strength printed on the label is a declaration with a margin set by law. In the European Union the margin is 0.5 % vol for wine and for beers up to 5.5 % vol, and wider for other categories.
Tolerances for wine
For grapevine products the rule is in Article 44 of Commission Delegated Regulation (EU) 2019/33. The actual alcoholic strength is indicated in units or half units of percentage by volume, followed by the symbol "% vol". Without prejudice to the tolerances of the reference analysis method, the indicated value may not differ, above or below, from the value determined by analysis beyond the following limits.
General rule
Maximum difference between the strength indicated on the label and the strength determined by analysis.
Categories with a wider margin
PDO and PGI products stored in bottles for more than three years, sparkling wines, quality sparkling wines, aerated sparkling wines, semi-sparkling and aerated semi-sparkling wines, liqueur wines, wines from overripe grapes.
Tolerances for beer
For beverages other than wine containing more than 1.2 % vol, the tolerances are in Annex XII of Regulation (EU) No 1169/2011 on the provision of food information to consumers. The strength is given to no more than one decimal place, followed by "% vol", and is determined at 20 °C. Here too the tolerances apply without prejudice to those of the analysis method.
± 0.5 % vol
Beers up to 5.5 % vol
± 1 % vol
Beers above 5.5 % vol
20 °C
Temperature at which strength is determined
What the measurements show
The Liquor Control Board of Ontario (LCBO), in Canada, tests every wine imported into the province and records its actual and stated alcohol content. Julian Alston, an economist at the University of California, Davis, and four co-authors examined 91,432 of these analyses, covering the years 1992 to 2009. Across all samples the actual strength exceeded the stated one by 0.13 points on average, but the average combines different groups.
Actual and stated strength in 91,432 wines
Groups defined by a difference of at least 0.01 percentage points between stated and measured strength. Average values in % vol.
| Group | Wines | Actual | Stated | Gap |
|---|---|---|---|---|
| Strength understated on the label | 52,178 (57.1 %) | 13.6 | 13.2 | 0.42 |
| Strength overstated on the label | 29,461 (32.2 %) | 12.9 | 13.2 | 0.32 |
| Strength correct within 0.01 points | 9,793 | 13.1 | 13.1 | 0.00 |
According to the authors, understatement was on average smaller in Old World wines, predominantly European, than in New World wines. The same authors point out that some tolerance is needed: labels may have to be printed months before the strength of the wine is finally determined, and one label may have to cover several lots of the same wine. The gaps they observed, they write, may be economically significant even when they stay within legal tolerances.
The study on UK beers, Science & Justice 2018
Peter Maskell and six co-authors used a near-infrared analyzer to measure the strength of 218 packaged beers, bought from a range of outlets across the United Kingdom over 12 months, with labeled strengths from 3.4 to 14 % v/v. The root mean square error between declared and measured strength was ±0.43 % v/v for beers up to 5.5 % v/v, ±0.53 % v/v for stronger beers and ±0.48 % v/v overall. The study comes from forensic science: the authors conclude that the variation is larger than the one used so far in blood alcohol calculations with the Widmark equation.
Strength before bottling
In wine and beer, density depends on alcohol but also on sugars, extract and other substances, so with the densimetric method alcoholic strength cannot be derived from the product as is. The OIV methods for wine require distillation of the sample and measurement of the density of the distillate, a mixture of water and alcohol, from which the strength is calculated with the alcoholometric tables. For beer, alcohol is determined in the same way on the distillate, with calculations per ASBC and MEBAK.
MicroDens
The MicroDens portable density meter by Gibertini Elettronica measures the density of the distillate and calculates alcoholic strength with the OIML R 22 tables: the Ethanol method for wine, Beer DryExt for beer, with the second reading on the distillate. Its uses in wine include pre-bottling control, the check of alcoholic strength against the values declared on the label; in beer, the determination of alcohol and real extract before packaging.
0.1 % v/v
Alcoholic strength accuracy
±0.0001 g/cm³
Density accuracy
1 mL
Sample per filling
- Commission Delegated Regulation (EU) 2019/33 of 17 October 2018, OJ L 9, 11.1.2019: Article 44, actual alcoholic strength.
- Regulation (EU) No 1169/2011, food information to consumers: Annex XII, alcoholic strength.
- J. M. Alston, K. B. Fuller, J. T. Lapsley, G. Soleas, K. P. Tumber, "Vinum Verum Viribus? Systematic Errors in Wine Alcohol Labels", ARE Update 19(4), 2016, Giannini Foundation, University of California. Summary of "Splendide Mendax", Journal of Wine Economics 10(3):275–313, 2015.
- P. D. Maskell et al., "The influence of alcohol content variation in UK packaged beers on the uncertainty of calculations using the Widmark equation", Science & Justice 58(4):271–275, 2018.
Last updated: November 12, 2026