AAWE Papers in Portland: Wine Price and Subjective Appreciation
Suppose good X is similar to good Y, but has a
considerably higher price tag. Most of us would probably expect good X to be
slightly better than good Y along
some dimension. Such an expectation would not imply that we would always
purchase good X: the difference in how much we appreciate the two goods may be
small, and not worth the increased expense. The point is simply that it seems
reasonable to expect there to be some
noticeable additional quality to the more expensive good.In the forthcoming issue of the Journal of Wine Economics, we present
our study “Do More Expensive Wines Taste Better? Evidence from a Large Sample
of Blind Tastings” (joint work with A. Dreber, J. W. Emerson, A. Herschkowitsch,
and J. Katz). Our main finding is that participants in blind tastings do not appreciate expensive wines more
than cheap wines. In
our sample of over 6,000 blind tasting observations, compiled by food and wine critic Robin
Goldstein and discussed at length in his new book, The Wine Trials (Fearless Critic Media, 2008), we find that the correlation between price and overall rating
is in fact small and negative,
suggesting that individuals on average enjoy more expensive wines slightly less.
There is no reason to expect individuals with a great deal of experience
of drinking wine to be similar to individuals with little experience in this
regard. Using wine training (such as participation in a sommelier course) as a
proxy for wine expertise, we were able to separate these “experts” from the
rest of the tasters in the sample. We find indications of a small positive
correlation for experts. This correlation is only marginally significant (the p-value is just below 0.10, despite the
large sample size) and the coefficient is quite small, suggesting that even for
expert tasters the correlation between price and subjective appreciation is
less than overwhelming.
In both cases, the coefficients are of a
moderate magnitude, but non-negligible, given that wine prices cover such a
large range. Suppose that Wine X costs ten times more then Wine Y in dollar
terms. In terms of a 100-point scale, such as that used by Wine Spectator, our estimates predict that non-experts will assign
an overall rating that is four points lower for wine X, whereas experts will
assign an overall rating that is seven points higher.
Why, then, do everyday wine drinkers spend
money on expensive wines? The price tag itself may influence how much we
appreciate the good. This is in line with a familiar finding in marketing
research: increasing the price of a good sometimes increases the demand for the good for psychological reasons alone
(see for example Cialdini, 1998), by signaling that the good is covetable. Goldstein explores this effect in The Wine Trials, which examines the methods of modern
wine marketing, the undue deference of consumers toward numerical ratings in
industry publications such as Wine
Spectator, and the inconsistencies between controlled blind-tasting results
and those sorts of ratings, which exhibit a strong positive correlation with
price.
References:
Goldstein, R. (2008). The Wine Trials.
Plassmann, H, J O’Doherty, B Shiv and A Rangel (2008).
Marketing Actions Can Modulate Neural Representations of Experienced
Pleasantness. Proceedings of the
Last I heard of this paper, it was still a working paper.
Has it undergone peer review prior to publication? Your conclusions seem unchanged.
Furthermore, your interpretation of the fMRI findings is incorrect.
In the paper itself, the authors acknowledge that while regions associated with "enjoyment" (but also suicidality, positive response to SSRIs, emotional inflexibility, etc) were more active in people who reported *enjoying* a particular wine more, the regions responsible for registering and recognizing the sensory information were not affected by suggested price.
This is more than a pedantic distinction. It has far reaching implications for understanding wine appreciation, as well people's ability to learn organoleptic assessment.
As for your definition of "expert" - it doesn't take much skill and information to obtain a basic Sommelier's Certificate or to become a Certified Wine Professional. You cannot take the possession of such a certificate as a guarantee of expertise.
That being said, I do not discount the fact that many discerning wine aficionados will recognize the quality and value of a relatively obscure, <$20 brand or variety.
At the end of the day, guess I am not 100% clear (or is it maybe, convinced) what the goals of this study were. What purposes will the conclusions of your publication serve? In what ways, can the conclusions and data be misinterpreted and misrepresented?
Reply to this
Arthur, I can't speak for the authors of this paper and will defer to them to reply to your questions/concerns. But I can confirm that the paper went through a regular peer review process. Sometimes the 'Journal of Wine Economics' can be fairly fast; maybe even most of the time.
Best, Karl
Reply to this
Arthur,
First, we are not sure where you get the idea that we make any claims about how the brain goes about "registering and recognizing the sensory information"--our paper doesn't mention those concepts. In our account of Plassman et al., we merely discuss activation in the orbitofrontal cortex, which is associated with enjoyment.
Second, sommelier certification and professional industry experience are widely accepted as proxies for wine expertise; if you propose that we measure wine expertise in a different way, it might help to suggest a viable alternative.
As for the goals of the paper, there are several, but one of the main ones is to raise awareness of the problems that arise when wine evaluations are done non-blind, as many are.
Reply to this
Johan,
In the piece above (credited to Robin Goldstein, one of the authors of your paper) it says:
"In a sense, our paper is a companion piece to Plassmann et al. (2008), in which experimental subjects reported higher levels of satisfaction from wine they were told was more expensive, even when it wasn’t; functional magnetic resonance imaging suggested that their brains responded differently to the taste experience itself."
That is not entirely correct. I have pointed out in past discussions elsewhere that your paper addresses similar concerns as the fMRI paper. But for most readers, the finer (and critical) points of that study get lost in the glare of the “gotcha” element. The main issue is that the region studied in the fMRI paper is not specific to enjoyment. Very few cortical regions actually are.
The main oversight of all the media hype over that study is that the authors indicated that there is a part of our brain function which is susceptible to external cues but is not related to the sensory input resulting from smelling/tasting the wine. It may be modulated by the former but it is not a slave to it. This fact is critical to the notion that people can develop skills by which they can independently and reliably judge wines beyond the scope of their own enjoyment of prejudices.
There are critics and experts who do just that. To discredit all wine critics across the board is not fair. But the undertones of the message are a greater disservice to consumers/enthusiasts because they indirectly support the false belief that there is no way (for critics or enthusiasts) to objectively, reliably and consistently judge wines.
With regard to “experts”: It is one thing to hold “participation in a sommelier course” (and apparently other training) as a proxy of expertise and another to use successful track records in the wine industry as measure of expertise. You and I both know that there are people who complete their economics or statistics training and don’t amount to much in their field. This is true for any field of study and any profession.
It is after reading the previously released version of the paper that I am concerned that your “expert’ cohort has many confounds. It has been some time since I read the working paper, but I believe that it seemed that the participating “experts” were identified as a matter of course during data collection and were not intentionally recruited into a specific cohort.
Reply to this
The second to last 'paragraph' shoud have read:
"That being said, I do not discount the fact that many discerning wine aficionados will recognize the quality and value of a well-made but obscure, <$20 brand or variety."
Sorry, my bad.
Reply to this