Michael Burry Is Buying Bordeaux to Short the Dollar. Here's the Math — and Why It's Hard to Copy

Last week, the investor who shorted U.S. housing before 2008 did something that has nothing to do with housing, chips, or crypto. He is buying First Growth Bordeaux — the top tier of French claret — and stowing most of it in bonded warehouses outside the United States. The headline framing sells it as "the best defense against dollar doom and AI disasters is a fine Bordeaux."

Strip the wine away and the trade is something drier. As Burry puts it, the move is "to take advantage of the reserve currency's remaining strength to buy cyclically depressed, novatable real assets outside its jurisdiction." That is not a lifestyle flex. It is a short position on the dollar that happens to be packaged as French wine. The real question for you is not whether great wine is a fine purchase. It is whether a bet structured this way is anything an ordinary investor can actually copy — and whether the numbers underneath hold up.

I think the honest answer is that it holds together as an idea and falls apart as a template. Let me show you the two pieces, because that's where your judgment has to go.

A "dollar short" wearing a wine label

Here is the mechanism, because it's the part most people miss. Fine wine held in a bonded warehouse in London or the EU is, in effect, an asset denominated in a foreign currency. Buy it while the dollar is still strong, and it becomes a claim priced in euros and pounds. When the dollar weakens, the wine rises in dollar terms — even if the wine itself doesn't move an inch. That is why Burry calls the position "a dollar short extraordinaire."

He lays out three legs. The first is timing: fine wine is in its deepest broad drawdown in the indexed era, with the London Liv-ex benchmarks down roughly 25 to 30 percent from their October 2022 peak. Buy a scarce asset near a cycle low and you're starting from a discount.

The second leg is the currency gain — and this is where the whole thesis either lives or dies. Burry expects the dollar to fall 30 to 40 percent over the next five to ten years as the fiscal bill comes due, and he estimates that decline alone would produce a 45 to 65 percent currency-driven gain on a wine held abroad. The third leg is the long tail: over twenty years he projects the wine itself could grow four to five times as the vintages age and their supply shrinks.

Notice the structure. The currency leg is the big one, and it is entirely conditional. It is not a fact about wine; it is a forecast about the U.S. dollar. If the dollar holds its ground, that 45 to 65 percent evaporates and you're left owning a 4 to 5 percent-a-year real asset that costs you money to store and is far harder to sell than an index fund. The wine is the delivery mechanism for a macro bet. The macro bet is the actual thesis.

The returns are real, but they are not the point

The best public evidence on what this asset actually earns is a 2015 study in the Journal of Financial Economics by Dimson, Rousseau, and Spaenjers, which examined more than 36,000 auction and dealer prices for the five First Growth Bordeaux from 1900 to 2012. The net real return — after estimated storage and insurance — was about 4.1 percent a year. For context, British equities returned 5.2 percent real over the same stretch. Wine outperformed stamps, art, and bonds, but it did not outperform stocks, and it carried ongoing costs to hold.

Burry's own math leans on that base and argues that buying 25 to 30 percent below a cycle peak resets the starting line, which he believes could push the long-run real return toward 6 to 8 percent. That is a fair argument. It is also a forecast, not a result.

There is a genuine structural case for why wine can work as a real asset, and it's worth stating plainly because it's the one part of the thesis that survives even without the dollar leg. Each bottle is genuinely different — region, producer, soil, vintage — so value builds over years rather than all at release. There is a deep, fragmented secondary market, which means prices are discoverable and you can actually buy at a discount to the prevailing level. And supply is permanently destroyed: every bottle that gets consumed is one that never comes back, so the finite stock of a great vintage only gets scarcer. That last point is what separates wine from a bourbon or a watch, where no bottle is ever consumed away.

Burry also makes a line I think is fair: gold already prices in "too much debt, too little monetary restraint," and you can't eat gold. Wine is a hard asset with a second use — you drink it.

His execution is disciplined, and it tells you who this is really for. He reviewed roughly 700 wines and bought only about 40, at average discounts of 18 to 20 percent. He set aside 15 to 20 percent of the purchase for personal consumption to offset the cost of owning the rest. And he caps the whole idea at under 10 percent of a portfolio, with a practical floor of five figures of net assets before it is even worth the bother.

The leg that's hardest to replicate is the entry

Here is the caution Burry offers that most headlines skip, and it's the single most important one for anyone tempted to follow. All of a top wine's appreciation can happen at the moment of winery release. He describes a dealer offering a top-rated 2010 Bordeaux below its release price — a bargain, on its face. Fifteen years later, in real terms, it had returned "less than nothing," because the value had already been captured by whoever bought it at release. You can have perfect wine and still own a dead investment. Buying well is the entire game, and the market is already showing early signs of recovery in 2026, which means the "25 to 30 percent below the peak" entry window is narrowing, not widening.

That's the honest bottom line. Burry's play has a real structural edge — a tangible, ageable, supply-shrinking asset with a liquid secondary market that is uncorrelated to your bank account, your pension, and your crypto wallet. But two of its three legs are forecasts, and the one leg that isn't (the depressed entry) is a timing bet that is already partly spent. For a person who genuinely believes U.S. fiscal debasement is likely, has the capital to size it properly, and can access bonded storage through a reputable dealer, a single-digit allocation is a defensible, if niche, hedge. For everyone else, it is an option on a macro outcome you would otherwise have to bet on directly — with higher friction, illiquidity, and cost than just holding cash or the defensive real assets it's meant to hedge. I would not reach for it as a first move. Understand what it is: a five-to-ten-year bet on the dollar, in a wine cellar, not a safe harbor you can copy on the way out the door.