Today's Wall Street Journal (WSJ) had a front page story on wine counterfeiting titled "U.S. investigates counterfeiting of rare wines" (WSJ online subscription required).  When I arrived at the office today, I had a voice mail from Robert Frank, the writer of The Wealth Report blog at the WSJ (which is billed as covering "the lives and culture of the wealthy").  He wanted to discuss how wine buyers could avoid fakes.  My comments are incorporated into his blog post today titled "Oenophiles see double."

In the discussion with Robert, I noted that buying wine sourced directly from the chateau (which is called "ex-chateau") or from individuals who bought it upon release is the ideal defense against fakes, as the "chain of provenance" is fully known (including storage conditions). 

Quantifying the value of buying ex-chateau 

I realized when making this point that I potentially had a method for quantifying the market pricing premium for authenticity. The other day when reviewing the price list of a very reputable British wine wholesaler (Farr Vintners) specializing in Bordeaux, I noticed that they listed certain Bordeaux wines with a notation that the wine was "ex-chateau."  If they happened to also offer the same wine, vintage, and bottle size without the ex-chateau designation (and in otherwise excellent condition), then an analysis of the price differences might reveal at least this firm's educated view of the price premium for authenticity (as storage conditions of non ex-chateau wine from Farr is typically impeccable).

The outcome 

In examining Farr's full Bordeaux price list, there were 18 examples that met the criteria which spanned the 1986-1998 vintage time frame including wines from Latour, Mouton-Rothschild, Haut Brion, and Pichon Lalande as well as less expensive Bordeaux such as Lafon Rochet.  The chart below displays the results with a trend line based on a linear regression approach.  A few observations:

1. The trend is for an increased premium for older wines which makes sense as the risk of fakes intuitively is greater the older the wine.
2. Only 1 of the 18 examples had no price premium (and this was for the cheapest wine).
3. Absolute price levels were less correlated with price premiums than vintage in this limited data set.  E.g., the five most expensive wines (priced at $2,000 to $5,000 a case), had 3 examples in the 9-10% premium range but also one at 5.6% and one at 2.8%.
4. The overall median premium in this sample was 5.7%.

One way of interpreting the authenticity premium is as a measure of the potential liability that an auction house or other seller of fine wine exposes itself when selling such wine.  If one extrapolated the trend line to 1947 (as an example of a great wine year), the premium (or potential liability of selling non ex-chateau wine) rises to 32.2%.  That's about the total gross margin percentage between buyer's premiums and seller's commissions that a major auction house would earn when selling a wine of that vintage.

Bottom line: Pay attention to provenance and be prepared to pay more for when it's solid -- it's worth it.