I have been traveling to Mexico and Brazil the past few years, and I have noticed that the wines are more expensive in Mexico compared to the United States, and substantively more expensive in Brazil. This is likely due to a variety of reasons, including supply and demand, the cost of logistics, and (maybe most of all) taxes.
So I decided to look up the average markup on wines around the world, depending on where the wine was from and where it was being sold, using Vivino data. I was curious if my observation of wines being extra expensive in Brazil was wrong or systemic.
To start, it’s helpful taking a look at specific cases. Take for example the famous Veuve Clicquot, the champagne that is sold all over the world. In its home country, France, Veuve Clicquot costs €41.95 (on median). Nowhere in the world does Veuve Clicquot cost more than in Thailand, where it costs €120.93 on median, a 188% markup from the price Veuve Clicquot sells for in France.
wdt_ID | Country of sale | Median price in France (EUR) | Median price in country sold (EUR) | Markup/ markdown |
---|---|---|---|---|
1 | Thailand | 41.95 | 120.93 | 188 |
2 | Peru | 41.95 | 105.34 | 151 |
3 | Brazil | 41.95 | 86.62 | 106 |
4 | Israel | 41.95 | 83.44 | 99 |
5 | Uruguay | 41.95 | 78.64 | 87 |
6 | China | 41.95 | 75.92 | 81 |
7 | Russia | 41.95 | 72.66 | 73 |
8 | Chile | 41.95 | 71.14 | 70 |
9 | Latvia | 41.95 | 63.99 | 53 |
10 | Ireland | 41.95 | 59.95 | 43 |
Country of sale | Median price in France (EUR) | Median price in country sold (EUR) | Markup/ markdown |
Taking this concept further, I then looked up all wines in the same way you see Veuve Clicquot above. I then took the median markup from origin country to foreign country across all the wines–the idea being, if Veuve Clicquot or any other wine for the matter over-indexed in markup, that error disappears in aggregate.
wdt_ID | Country of wine origin | Currency in origin country | Country of sale | Markup/markdown | Distinct wines in dataset |
---|---|---|---|---|---|
1 | South Africa | ZAR | Thailand | 425 | 36 |
2 | Portugal | EUR | Thailand | 410 | 22 |
3 | Italy | EUR | Uruguay | 275 | 18 |
4 | South Africa | ZAR | Malaysia | 272 | 65 |
5 | Germany | EUR | Brazil | 269 | 21 |
6 | Spain | EUR | Thailand | 258 | 130 |
7 | South Africa | ZAR | Mexico | 252 | 18 |
8 | South Africa | ZAR | Brazil | 245 | 91 |
9 | South Africa | ZAR | China | 222 | 41 |
10 | Germany | EUR | Thailand | 215 | 51 |
Country of wine origin | Currency in origin country | Country of sale | Markup/markdown | Distinct wines in dataset |
Sure enough, wines are more expensive in Mexico: the same American wine offered in the United States costs 71% more in Mexico, and Italian wines cost 102% more in Mexico compared to Italy.
Brazil is more of the same and more: the same American wine offered in the United States costs 156% more in Brazil, and Italian wines cost 201% more in Mexico compared to Italy. This markup is about double that of Mexico.
The most expensive markup in the world is South African wines sold in Thailand, where they cost 425% more than in South Africa. Thailand has a “sin tax” policy on goods like alcohol, cigarettes, and more. It’s the reason why Thailand is prevalent near the top of the table above.
On the markdown side are a few interesting cases. Greek wines sold in Germany cost 5% less than they do in their home country. Without investigating this, it makes sense on the surface: there are low logistic and tax barriers since both countries are in the EU. However, with the Greek economy struggling, the country is incentivized to drop the price of their goods to encourage an increase in exports.
Below are some methodology notes. Please comb through the data above and reach out if you see anything particularly interesting.
I took a look at at every wine’s median online price in its “home” country, based on listings from over 25,000 online merchants. I then looked at the median prices of these wines in individual foreign countries, also based on the 25,000 online merchants. I converted the price in a foreign market to the “home” currency, based on average currency conversions from May 2019.
I calculated the markup for every distinct instance of [wine, home country, and foreign country]. Markup is calculated as [foreign price]/[home price]-1. This is the Veuve Clicquot table above.
Finally, I calculated the median markup of all distinct instances of [home country, foreign country]. These instances needed to have at least 10 wines with a markup; the less instances of wines, the less trustworthy the data is. This is the country markup table above.
In summary, this analysis is a median of median markups comparing specific wine prices both in its domestic market and foreign markets.