Wine Pricing

[marcus]

Claude Kolm stated on the Ledoyen thread -- "Here in San Francisco, for example, and I am sure that it is not just here, buyers frequently pay substantially more for wines at auction than they would pay at stores for the same wines. One would think that wine-searcher.com would end such follies, but it is not the case."

As someone who purchases the majority of my wine at auction, I felt that I was essentially being told that I was engaging in folly and didn't know what I was doing. I decided to do my own comparison and the results are below. For this purpose I selected the 1982 first growth Bordeaux with the addition of Cheval Blanc. These are benchmark wines which trade frequently at auction and are widely available through fine retailers and so should best illustrate the dynamics of these two parallel markets.

For auction results I looked at Sotheby's New York sales 2002-2003 season only. I looked for the lowest price that I could quickly spot and checked the listing to make sure that there were no wine condition problems. If there were, I excluded the result. I did not check other auction houses or other US locations and I could have undoubtedly shaved the prices if I had done so. For wine-searcher, I picked the lowest retail price providing it did not show a condition problem. A significant number of listings on wine searcher are actually for ongoing auctions where the price listed is the latest bid, and one needs to be careful in order to avoid these. In the case where there is a wide gap between the lowest price on wine searcher and the next lowest, I show both.

Cheval Blanc $650 John Hart, Chicago $514 Sothebys 3/8/03

Latour $549 WH Frank, NY $441 Sothebys 11/22/02

Lafite $425 Premier Cru, CA ($499 next low) $367 Sothebys 09/14/02

Mouton $525 John Hart, Chicago $343 Sothebys 3/8/03

Margaux $495 Grapes, NY ($599 next low) $343 Sothebys 09/14/02

Haut Brion $325 Grapes, NY $215 Sothebys 3/8/03

The results speak for themselves.

Some additional comments:

There is far more provenance and condition information available from the auction houses than through the wine-searcher website.

I believe that far more of the high quality older wines trades through the auction market, rather than the small number of retailers that carry them. The auction market is thus more liquid and the prices more real. How many bottles do these retailers actually own in order to meet demand?

The spread of prices is much greater in the retail market than the auction market. The median price on wine searcher is much higher than the median auction price in relation to the low price. This implies that the ratio of the median retail price to the median auction price is much greater than the ratio of the low prices. The ratio of medians is probably an even more important indicator of the pricing relationship of these two markets.

[Claude Kolm/The Fine Wine Review]

Marcus -- I did not mean to imply that you personally were engaging in folly. I only meant that one sometimes sees prices that are signifcantly above the costs of the same wines in stores -- that does not imply that one always pays more at auction.

A question for you and then a few of observations:

Do the auction prices that you quote include the buyer's premium? If not, you really should adjust your figures to add that in, in which case the results are closer.

You are comparing individual bottle prices of very expensive wines with (presumably) individual bottle price allocations of wines sold in lots of a case or more. When wines get this expensive, it is possible that there is a significant reduction for case lots.

The prices on wine-searcher.com surprise me. Not long ago (just before Christmas if I recall correctly), 1982 Mouton was selling here in the Bay Area for $425-499, and now the cheapest that you can find it for is $525.

Finally, while the results are interesting, they are limited to wines from a single year, and a year where for whatever reason there seem to be significantly fewer first growths in the stores than in the recent past. For example, the only 1982 first growth at Premier Cru at the moment is the Lafite that you quote, and the only one at K and L is Petrus. Usually, both stores have more in stock -- suggesting that something may be going on with 1982 first growths.

Best regards,

Claude Kolm

The Fine Wine Review

[marcus]

Claude -- Of course the buyer's premium is included, I should have mentioned that in my post. The hammer price is essentially meaningless.

With regard to the number of bottles in a lot, I haven't observed that this makes a significant difference in price. Odd size bottles, which I didn't include in my comparison, typically do sell at a premium because they are rarer. Half bottles of older wines sell at large premiums, and magnums at about a 5-10% premium. My own preference is to buy at least 6 bottles, preferably a case, so one can get to know a wine over time. One problem that I expect one would find is that the retailers on wine-searcher may in fact have only a few bottles, not as many as you might like to buy. If that were in fact the case, it would truly not be comparable.

I believe that I could go through this exercise for any year up through about 1996 and get similar results. By its fundamental nature, a liquid auction market will almost always yield average prices significantly lower than a retail market. For vintages later than 1996, there would be the overhang of the original release and the original wholesale pricing would still be a market factor. I chose 1982, without any foreknowledge, because it is the key current benchmark year for Bordeaux and I felt that it would be most representative. I still see these 1982s in local NYC stores.

[Claude Kolm/The Fine Wine Review]

Marcus -- I checked your wines against the prices at the last Butterfield's sale here in San Francisco (March 17, 2003). (www.butterfields.com) It is unclear on the web page whether the prices are the bid price or the purchase price (which includes the 15% buyer's premium) -- but I assumed the former and added the 15% on. The results are as follows:

Lafite: 3 lots sold, ranging from $468.39 to $477.57/bottle which is above the current Premier Cru price.

Mouton: 4 lots sold, ranging from $440.83 to $523.59/bottle, which is just below the John Hart price you quote.

Latour: 1 lot at $523.29/bottle

Margaux: 7 lots ranging from $275.52 to $413.28/bottle

Haut-Brion: 6 lots ranging from $247.96 to $358.17/bottle which is above the price that you quote for Grapes, NY

Cheval-Blanc: 1 lot at $606/bottle.

Additionally, Pichon-Lalande caught my eye. Six lots of it sold for $275 - $330/bottle. Wine-searcher gives a lowest price of $299.50, but in fact if you go to Premier Cru's website, you will see that they are selling it for $249.50 (interesting, since they are one of the firms listed on Wine-searcher.

So bottom line is, some people at auction do better than they could in a store, but as I claimed, some do not. Moreover, given the current softness of the market (at least out here), I'm pretty sure that one can do some bargaining with stores (if one can locate stores with the inventory).

Best regards,

Claude Kolm

The Fine Wine Review

[marcus]

Claude -- I took a quick look at Butterfield's price list and it is clear that the buyer's commission is included in the price. These prices you listed are all 15% above standard bidding intervals. If you adjust your prices, they will be largely inline with those that I found. Also, for fair comparison, you should only be referencing the least expensive auction price, because if you were also to show the range on wine-searcher, the high end of that range would be humongous.

With regard to Pichon Lalande, there was one lot at Sotheby's that sold for $235/btl, but most tended towards $260. Premier Cru's price of $249.95 is good because its in the ballpark of the auction price, its not really better. Also, how many bottles do they really have to sell. The median price among the sellers of this wine on wine-searche is $375 which is well above the auction price.

[Claude Kolm/The Fine Wine Review]

Marcus -- I had overlooked the fact that you were only reporting the lowest auction prices. The results of that search are irrelevant to the point that I was making -- all it shows is that some bidders can beat the price that they could have purchased the wine for elsewhere. But the point that I was making is that some (other) bidders do not beat the price that they could have purchased the wine for elsewhere. To rebut my point, you would have to look at the maximum sales prices at auction.

In other words, the point that I made in the Ledoyen thread is that the well-informed bidder should know what the wine can be bought for elsewhere is and not make any bids that would result in a purchase price above that, and some people fail to do that.

Best regards,

Claude Kolm

The Fine Wine Review

Edited April 11, 2003 by Claude Kolm/The Fine Wine Review (log)

[marcus]

Claude -- I think that I made the mistake of thinking that you were intending to make a relevant point about relative values available in the auction market, versus the retail market as enhanced by internet comparison capabilties. The fact that different people pay vastly different prices for the same item, regardless of the market in which it is purchased, is obvious and trivial and something that we all know already.