The new issue of the Journal of Wine Economics (JWE) leads off with a fascinating article by Julian M. Alston, Kate B. Fuller, James T. Lapsley and George Soleas titled “Too Much of a Good Thing? Causes and Consequences of Increases in Sugar Content of California Wine Grapes.”  It’s the kind of wine economics research that makes me smile: rigorous, clever, relevant — and it even includes a poem! What could be better!

Demand-Side Explanations: The Parker Theory

The puzzle that the article examines is why sugar levels (measured in degrees Brix) have risen in California, taking wine alcohol levels with them. Sugar levels in white grapes grown in California increased 12% from 1980-84 to 2005-08, for example, with the average degree Brix rising from 20.7 to 23.2. Average Brix for red grape varieties increased from 22.2 degrees Brix to 24.3. Sugar levels in Cabernet Sauvignon grapes increased from 22.8 to 25.0 degrees Brix at harvest.  Higher sugar levels mean higher alcohol levels, all else being equal.

Two simple explanations are usually cited to explain the rising sugar/alcohol trend. The first is based on changes in demand. Robert Parker (and some other powerful wine critics) are said to prefer a certain style of wine that is riper. Sugar and alcohol levels have increased as wine growers have worked to produce the grapes that make the wines that most please the Golden Palate of Robert P. (or that otherwise appeal to changing consumer preferences).

Supply-Side Explanations: Chateau Al Gore

I call the second theory the “Chateau Al Gore” hypothesis because Al Gore is associated in popular culture with global climate change and that’s what this theory is about. Rising temperatures, according to this line of reasoning, produce riper grapes pushing up sugar levels, boosting alcohol.

It is pretty easy to line up facts to make a persuasive case for Chateau Al Gore. Temperatures as measured by a heat index have been rising in California, according to the article’s authors. Sugar and alcohol levels have increased, too. Although additional sugar may be welcome (the Parker principle), there are indications that producers would prefer lower levels. A good deal of wine in California is partially de-alcoholized, for example. Alcohol is removed from a portion of the vintage (using reverse osmosis or the spinning cone method I am told) and then the treated wine is blended back, reducing overall alcohol levels and allowing winemakers the opportunity to find the “sweet spot” alcohol level for their wines.

Some of the de-alcoholization may be motivated by federal taxes, which increase substantially on a per-gallon basis for wines that rise above the 14% ABV level. The extra 50 cents tax per gallon may not concern the makers of expensive wines like Screaming Eagle, but it can be a significant cost factor for bulk producters and thus worth the expense of alcohol reduction. In any case, the authors find that lower-priced grapes tend to have lower average Brix readings, which is consistent with the tax hypothesis but doesn’t prove it.

If alcohol levels of wine have increased even after partial dealcoholization, this suggests that rising sugar levels must be unwanted and this notion is at least partially confirmed by preliminary data reported here that many wineries under-estimate alcohol levels on product labels. The authors have obtained access to data from the Liquor Control Board of Ontario (one of the largest wine merchants in the world), which show that the average stated alcohol level of California wine exported to Ontario was 12.63 percent in the sample period while the actual level was 13.35 percent. The gap is clear, but the authors suggest that more work is needed to fully understand it. I’ll be interested to read their final report.

Ceteris Paribus

The Chateau Al Gore theory seems pretty persuasive. Ceteris paribus (holding everything else constant) it makes sense that sugar and alcohol levels would rise with average vineyard temperature. The fact that winemakers work to offset the alcohol boost and maybe fudge it a bit (within legal limits)  on the label suggests that this is a climate change event that they struggle to contain.

But ceteris is seldom really paribus in wine. Employing multi-factor regression analysis, the authors find that the rising heat index is responsible for some of the increase in sugar levels, but not all of it. Put another way, climate factors alone are not sufficient to explain the total increase in sugar and alcohol. Other factors must also be at work.

Which pushes us back to the demand-side Parker Theory, but in a  usefully complicated way. It is important to understand how much the California wine industry has changed in the last 30 years. The type of wine produced has changed, for example, with varietal wine (Chardonnay, Cabernet Sauvignon) increasing faster than production of generic wine (“Chablis,” “Burgundy”). Varietal wines accounted for 71% of California production in 2000, according to the authors, compared to just 19% in 1985.

The move upmarket required different grape varieties of higher quality from different production zones. Thus while total California wine grape production rose by 60% in the survey period, the biggest increase (+185%) was in the Delta region (including the Lodi AVA) with the North Coast (including Napa and Sonoma Valleys) increasing by 128%.  Wine grape production in the San Joaquin Valley rose but by a much smaller amount. The southern San Joaquin valley accounted for just 30% of California vineyard acreage in 2008 (down from 50% in 1981), although it still produced more than 60% of wine grape tonnage because of higher yields.

So wine grape production has increased and also shifted in terms of desired quality, price per ton, grape variety and growing location. It is perhaps not surprising that average sugar levels would change too. Much of the growth in the California wine industry has thus been associated with the demand shift towards premium and ultra-premium wines and the rising sugar levels are to some extent associated with this “grape transformation” of the American palate. Robert Parker is part of this movement although I think it would be unfair to give him all the credit or blame for the changes noted here.

Getting to the Root of the Problem

So the JWE article finds evidence for both the demand-side and supply-side theories of rising sugar levels. But wait, there is more, including rootstocks (this is the “real dirt” in the title of this post).

Phylloxera struck California starting in the mid-1980s when the supposedly Phylloxera resistant AxR#1 rootstock was found to be susceptible to this root-sucking parasite. Eventually nearly all the vines affected were grubbed up and replaced with vines grafted to different (and hopefully more resistant) rootstocks. Several winemakers have suggested to me that the new rootstocks and associated changes in viticultural practices affect grape ripening — sugar levels peak before the desired flavor profile (phenolic ripeness) has been achieved. Longer hang times are needed to get the flavors right,  leaving wine growers with the problem of too much sugar and so forth.

The rootstock hypothesis is beyond the scope of the JWE study, but it indicates how complicated it can be to explain seemingly simple questions in wine economics and how much wine remains an agricultural product after all.

>>><<<

I don’t think I’ve done justice to this research so I hope you will click on the link in the first paragraph above and read the study yourself.