The Proximity Principle of Wine: How Neighboring Wineries Affect the Market Price of Wines in a Region

“One bad wine in the valley is bad for every winery in the valley. One good wine in the valley is good for everyone.” –Robert Mondavi

Leading in to the article presented today (Yang et al. 2012) was the above quote by Robert Mondavi in regards to the Napa Valley in the 1960s. It’s pretty self-explanatory in that he and many others believe that one bad winery in an area will decrease the attractiveness of all other wineries in the region, and one good winery in the area will increase the attractiveness of all other wineries in that region. Past research has found that the location of a particular winery will have a significant influence on the market prices of those wines and subsequently those wines in the immediate area. Consumers look for recognizable locations on a label when purchasing wines, as well as the price, expert opinions of the wine, and quality.

There are many reasons Yang et al. (2012) cited as to why it is beneficial for wineries to be in close proximity to each other. A winery that is geographically isolated may suffer from the fact that they may not be in an appellation or wine region that consumers recognize. Also, being geographically isolated means that visitor attraction may be decreased, since consumers tend to prefer visiting several wineries in a short span of time instead of visiting only one. Other benefits include potentially lower production costs, as labor is often shared between wineries in close proximity to one another.

Yang et al. (2012) also cited a few reasons why spatial clustering of wineries could negatively affect the price and reputation of the wineries in that area. First, pests or disease pathogens could easily spread from one winery or vineyard to another when in close proximity, thus damaging or destroying the crop from multiple wineries in the area. A winery in isolation is less likely to be infected by this type of pest or pathogen outbreak. Also, a winery focused on producing cheap mass-produced wines could take advantage of the close proximity principle and move into an area with higher quality wines, thus artificially increasing their own prices and increasing their initial chances of being purchased by a consumer, simply by having a recognizable region name on the label. Land prices also tend to be significantly higher in an area with higher quality wines, thus making it more difficult for new wineries to become established.

Taking both the positive and negative effects together ultimately determines how a particular wine region will develop over time. Specifically, these factors influence how wineries will establish themselves spatially, as well as the quality and price of the wines produced in that area. Though it has been generally shown that location affects price of wine, according to Yang et al. (2012), the specific spatial effects of the economics of wine production has been largely unexplored.

The paper presented today aimed to use knowledge obtained from spatial hedonic literature as well as knowledge obtained from hedonic modeling of consumer goods literature and created a spatial econometric model to determine how geographic proximity affects economic relationships among

By Matt Pourney [Public domain], via Wikimedia Commons

By Matt Pourney [Public domain], via Wikimedia Commons

wineries as well as the general reputation of a wine region. Using GIS (geographic information system) data and spatial analyses of the California and Washington wine regions, the authors aimed to get a better understanding of how economic factors as well as other relationships are influenced by geographical location.


The data used in this analysis was for wineries that produced red wines in Washington and California. For each wine, data on price, scores, case numbers produced, years of aging, vintage, and the production region were collected. Data for each individual winery was pooled to obtain one value per winery.

Wine regions for California that were used in the analysis were: Napa, Sonoma, Mendocino, Bay Area, South Coast, Carneros, Sierra Foothills, and “other” California. Wine regions for Washington that were used in the analysis were: Columbia Valley, Yakima Valley, Walla Walla, Puget Sound, and “other” Washington. A total of 79 wineries were analyzed for Washington and 876 wineries for California. There were more wineries in each area, however, exact location information was not available at the time and were thus not included in analysis.

All information included winery names and addresses were collected from the Wine Spectator online site. Addresses were converted to latitude and longitude GIS coordinates. It was assumed that wine production took place at the address provided by the Wine Spectator database. Since winery information came from the Wine Spectator, the authors acknowledged that the quality limits for wineries used in the analysis were slightly higher than what they may be in reality, as there were no boxed wine or jug wine producing wineries included in the analysis.

Two mathematical models were created for this spatial analysis. The first model was a hedonic price model with a spatial lag parameter which included the following variables: wine score, number of cases produced by the winery, the number of year the wine was aged before release, and the wine region. The second model examined the spatial correlation in wine prices independent from any other variable.


• Modeling results showed that close neighbors have a significant and positive effect on the price of a winery’s own wine.
o Higher-performing wineries have a significant positive impact not only on their own wine prices, but also the prices of their close neighbor’s wine prices. This was true for both Washington and California wineries.
o Wine scores had a significant and positive effect on wine price. (The higher the score, the higher the price).
o The number of cases produced had a significant and negative effect on wine price. (The more cases produced the lower the price).
o The number of years the wine aged before release had a significant and positive effect on wine price. (The longer the aging, the higher the price).
o Specific wine region did not affect wine prices.
o Wine scores had a significant and positive effect on wine price. (The higher the score, the higher the price).
o The number of cases produced had a significant and negative effect on wine price. (The more cases produced the lower the price).
o Specific wine region significantly affected wine prices. Specifically, the model showed a clear benefit to being in the Napa or Sonoma wine regions (i.e. higher prices in these areas).


According to the authors, this is the first study to use GIS to evaluate the effects of geographic proximity of wineries on wine prices. Their results suggest that there is, in fact, a clustering effect of wineries on the market price of wines in that area, and that the interactions between the wineries in an area are very important in regards to price. It was clear from the results that expert ratings and scores positively affected price (i.e. high score = higher price), and that being located in a more recognized region (i.e. Napa or Sonoma) increased the prices of all wines in that area. It was interesting to note that there was no region effect on prices in Washington state, the reasoning of which was not explored in this paper but may be due to the fact that the region is younger and in general smaller than in California, so brand awareness occurs more on a state-wide level instead of a region-wide level (my theory anyway).

While the results of this study are fascinating, it brings up many more questions that were not addressed and which could make interesting follow-up studies. For

By Agne27 (Own work) [CC-BY-SA-3.0 ( or GFDL (], via Wikimedia Commons

By Agne27 (Own work) [CC-BY-SA-3.0 ( or GFDL (], via Wikimedia Commons

example: what are the specific factors or reasons that result in this neighborhood effect? Is it the stories behind the region (i.e. history, romance, etc), or is it the specific terroir and style of the region that has people willing to spend more money?

Also, what happens in the long term when a winery producing lower quality wines tries to take advantage of the good neighbor effect and moves into a region with a lot of higher quality wineries? The authors of this study suggested that there is the incentive for these types of wineries to move into these regions, but may ultimately undermine the area by lower the overall reputation of the area over time. Understanding how these spatial effects equilibrate themselves over the long term would be very fascinating to explore.

Finally, the authors noted that they did not take costs into consideration when developing this model. Would it be economically beneficial for a winery to move into an already established wine region such as Napa or Sonoma where land prices and other costs are markedly higher than in lesser known regions? Do the benefits received from the good neighbor effect outweigh the negative cost effects? Or would it be more economically beneficial to set up camp in a lesser known region that costs less though may not receive as much traffic or as much purchases from consumers since it is a lesser known region? It would be very interesting to see what would happen to the model if these cost factors were taken into consideration.

Overall, the results of this study provide evidence for the close proximity theory for wineries, and that it is beneficial to be close to wineries producing high quality wines. The information found in this study could be helpful for those that are considering investing in a winery in Washington or California, however, future research would need to be done, particularly in regards to costs, in order to be able to make a fully-informed decision on whether or not to set up a winery in a particular region or not.

I’d love to hear your thoughts on this topic! Please feel free to comment and share!

Source: Yang, N., McCluskey, J.J., and Brady, M.P. 2012. The Value of Good Neighbors: A Spatial Analysis of the California and Washington State Wine Industries. Land Economics 88(4): 674-684.

6 comments for “The Proximity Principle of Wine: How Neighboring Wineries Affect the Market Price of Wines in a Region

  1. Andrew Waterhouse
    April 1, 2013 at 10:29 am

    This is a very important observation. It is the basis for much greater cooperation between wine companies compared to other businesses. Those wine companies that are not cooperating with their neighbors should read this. They are hurting their own bottom line!
    I would like to see if there is an effect on land prices too.

    • Becca
      April 2, 2013 at 7:27 am

      Thanks for your comments, Andrew! It’s always been assumed that neighboring wineries help each other (or at least some assumed this!), but seeing published research supporting the theory I would hope would encourage wineries to work together a little more.

      I agree it would definitely be interesting to see if there is an effect on land prices….I’d be willing to bet there is!

  2. April 2, 2013 at 10:46 pm

    It is very unfortunate that one rotten apple can ruin the barrel. Wine companies can’t influence each others’ wine within a region, so like the rest of us, they need to hope for the right neighbor.
    On the whole, though, those observations are not surprising, and reflect everything that I have seen and experienced.

    • Becca
      April 3, 2013 at 9:16 pm

      Thanks for your comments, Diana! Like you, I’ve seen these observations in action, and am glad to see it confirmed in a research journal!

  3. April 16, 2013 at 5:38 pm

    Correlation does not equate to cause. Just because high scored wines cost more does not mean that the score “raised” the price. It could just as easily be the reverse (a high price leads to a high score) or countless other factors. Even scored fully blind (which rarely happens), critics can still taste the money (better sorting, barrels, technical tweeking) that went into the wine and typically rate those better.

    • Becca
      April 16, 2013 at 6:20 pm

      You’re completely right, Aaron! Correlation definitely does not equal cause!

      Thanks for your great comments!

Comments are closed.