Wines & Vines

October 2016 Bottles and Labels Issue

Issue link: http://winesandvines.uberflip.com/i/732978

Contents of this Issue

Navigation

Page 55 of 83

56 WINES&VINES October 2016 WINEMAKING PRACTICAL WINERY & VINEYARD E very wine imported into Ontario, Canada, is tested by the Liquor Control Board of Ontario (LCBO). The LCBO records several characteristics of each wine, including the actual and stated al- cohol content. We analyzed data on 91,432 individual wines tested by the LCBO. Our analysis yielded two main findings: First, between 1992 and 2009, the alcohol content in wines trended up, with an overall average in- crease of about 0.5% alcohol by volume (ABV) on a base of 12% to 13% ABV. Second, the actual alcohol content differed systematically from what was reported on wine labels, with labels tending to understate the alcohol content for higher alcohol wines and overstate the alcohol content for lower alcohol wines. For example, of the 14,218 LCBO wines imported from California, 8,880 (62.5%) understated the alcohol content on their labels. Generous legal tolerances for labeling errors These labeling errors may be economically significant, even if they do not exceed legal tolerances. While every bottle of wine re- ports alcohol content on the label, the toler- ances are wide. U.S. law allows a range of plus or minus 1.5% for wine with 14% ABV or less and plus or minus 1% for wine with more than 14% ABV. Other countries have similarly large tolerances. These are wide bands compared with the relevant range of variation in the marketplace—the vast ma- jority of wine consumed as table wine has between 12% and 15% alcohol. The tolerance for labeling errors does not permit misclassifying wine between tax cat- egories in the United States. It is not legal to label wine as having more than 14% alcohol if it has 14% or less, and it is not legal to label wine as having 14% or less if it has more than 14% alcohol; in practice, 14.04% is the upper limit, reflecting some tolerance for measure- ment error (see page 86 of ttb.gov/pdf/com- pliance-seminar.pdf). Our analysis includes test results for 16,184 U.S. wines, among which 1,412 (9%) are labeled as having 14% alcohol or less when the actual ABV is more than 14%, and 299 (2%) are labeled as having more than 14% alcohol when the ABV is actually 14% or less. It would be illegal to sell these wines with those labels in the United States. Some tolerance for error in wine alcohol labels is appropriate for several reasons: One reason is to allow for measurement error, since the instruments (and perhaps their users) are not always perfectly precise. In addition, commercial reality dictates some tolerance to allow for the fact that wine labels may be printed months in advance of the final determination of each particular wine's con- tent and therefore must be based on predic- tions made when the labels are ordered. A particular label may also have to apply to multiple lots of the "same" wine that vary in their alcohol content. Even so, a tolerance of +/– 1.5% ABV seems generous. What do winemakers know about the alcohol content of their wine? To comply with tax regulations (at least in the United States), it is necessary to provide infor- mation about alcohol content. The U.S. tax rate is $1.07 per gallon for wine with 14% alcohol or less and $1.57 per gallon for wine above 14% but less than 21% alcohol. More important, alcohol content is also an element of quality control in winemaking. Con- sequently, we expect that commercial wineries, for the most part, have reasonably precise knowledge of the alcohol content of the wines they produce. It is relatively inexpensive to measure the alcohol content of wine reasonably precisely, although some of the devices used may entail larger measurement errors. The most common measurement method for wineries is the ebulliometer, which compares the boiling temperature of water and the boiling point of wine to determine alcohol concentration (which lowers the boiling point). This method has a reported accuracy of +/– 0.2% ABV. Ran- dom measurement error at this rate cannot ac- count for the patterns recorded in our data. On average, across all of the wines tested, the labels understated the actual alcohol content by 0.13% ABV. In other words, the Don't Believe Everything You Read Accounting for errors in stated alcohol content By Julian M. Alston, Kate B. Fuller and James T. Lapsley

Articles in this issue

Links on this page

Archives of this issue

view archives of Wines & Vines - October 2016 Bottles and Labels Issue