Investing in Bordeaux

On a Tuesday morning in late October 2010, a case of Château Lafite Rothschild 2008 was trading on Liv-ex at £8,500. By Wednesday lunchtime it was £10,160. Nothing had happened to the wine, still in barrel in Pauillac. The château had announced that the 2008 bottles would carry the Chinese character for eight, a lucky number in the market that had become Lafite’s largest, and a First Growth had gained nearly 20% in a day on a label design (Decanter, October 2010). Days later, on October 29, Sotheby’s sold almost 2,000 bottles straight from the Lafite cellars in Hong Kong, and three bottles of the 1869 went for US$233,972 each, at the time the most ever paid for a bottle of wine at auction (Sotheby’s, Decanter, Bloomberg).

That was the top. In June 2011 the Liv-ex Fine Wine 50, which tracks the ten most recent physical vintages of the five First Growths and had risen 113.9% in the 2009–11 bull run, printed 448.38 and turned down. By January 2012 it had given back 47.6% of its entire rise, and by 2014 it had fallen all the way to its 2008 peak, a 76.4% retracement of the China boom (Liv-ex). Some vintages of Lafite lost a quarter of their value inside six months (Liv-ex, November 2011), and Bordeaux spent the next five years out of favour.

Eleven years after that peak, the market did the same thing more slowly. The Liv-ex Fine Wine 100 peaked in September 2022 and spent the next 35 months falling, to a low in August 2025 (our tape). Three en primeur campaigns in a row, for the 2023, 2024 and 2025 vintages, asked buyers to commit money to wine in barrel; the third closed in June 2026 with what Liv-ex’s own closing report called demand for “only a handful of wines”. This guide is about what is left when the marketing is stripped away: a 170-year-old ranking, five estates that make 100,000 to 300,000 bottles of their grand vin a year, a 17th-century distribution chain, three complete price cycles, and a set of rules for buying at a price that can still make money.

Bordeaux as an investment is not a region; it is a list. In 1855, for the Paris Exposition, the city’s wine brokers ranked the Médoc’s châteaux by the prices their wines had been fetching, and sorted them into five tiers, or growths. The list covers 61 red-wine estates, all from the Médoc except Château Haut-Brion in Graves, and in 170 years it has changed once: Mouton Rothschild was promoted from second to first growth in 1973. No other asset class prices itself off a Second Empire list that nobody may update, and that permanence is what makes the top of it investable: a buyer in Hong Kong, London or New York knows what a First Growth is without being told. That shared knowledge is the source of Bordeaux’s liquidity.

The five First Growths, Lafite Rothschild, Latour, Margaux, Haut-Brion and Mouton Rothschild, are the mega-caps. They are also, by the standards of fine wine, large producers, and that matters more than a newcomer expects. Lafite makes 35,000 to 40,000 cases a year across its two wines, of which the grand vin is roughly 40%: between 180,000 and 300,000 bottles of Château Lafite Rothschild itself, depending on the vintage (producer and merchant data via Vinovest, Bid for Wine and FICOFI).

Latour’s grand vin runs about 220,000 bottles, Margaux about 150,000, and Haut-Brion, the smallest, 7,000 to 10,000 cases (merchant and producer references); Mouton’s figure was not re-verified for this guide. Multiply that by the 40 vintages still drinking and there are millions of bottles of each in circulation, against a few thousand bottles a year of Romanée-Conti, which is why a case of Lafite can be sold in days and a case of Burgundy cannot always be sold at all. The scarcity end of the trade is the subject of the sister guide, Investing in Burgundy and Champagne; this one stays on the liquid end.

Beneath the First Growths sits the tier the trade calls the super-seconds: Léoville Las Cases, the two Pichons, Cos d’Estournel, Ducru-Beaucaillou, Montrose and Palmer, second and third growths on the 1855 list whose prices run close to the first tier in strong vintages. They are where the value has historically been, and where liquidity thins once you leave the best-known six or seven names.

The Right Bank, Pomerol and Saint-Émilion across the Dordogne, is Merlot country and was never classified in 1855. Pomerol has no classification at all, which has not stopped Pétrus and Le Pin from trading above every First Growth; Le Pin makes roughly 400 to 600 cases a year (Sotheby’s). Saint-Émilion runs its own classification, revised roughly every decade, and in 2022 it fractured: Cheval Blanc and Ausone, two of the four estates ranked Premier Grand Cru Classé A, withdrew in July 2021 rather than be judged on criteria that, for the 2022 ranking, weighed marketing, tourism and architecture alongside the wine, and Angélus followed (Decanter, Wine-Searcher, Wine Spectator). None of them lost anything in price. On the Right Bank the classification is a footnote and the name is the asset.

The wider region is in a different business altogether. Bordeaux has about 94,700 hectares under vine after a subsidised grubbing-up scheme removed 9,500 hectares, an 8% cut that took the vineyard to its smallest in almost 30 years; the scheme was funded at €57M in March 2023 and has since been extended, with 12,263 hectares expected to be pulled by the time it ends (CIVB via Decanter, The Drinks Business, Jane Anson and Wine-Intelligence). The 2024 and 2025 crops both came in about 15% below the long-run average, and exports in the 2024–25 wine year fell 7% to 3.25M hectolitres (Bordeaux Wine Vacations 2025 report). None of that touches the 60 or so estates that trade on Liv-ex: the investable universe is a rounding error of Bordeaux’s output, and it is getting no larger.

The long-run number for Bordeaux is the long-run number for fine wine, because the only century-long series anyone has built is a Bordeaux series. Dimson, Rousseau and Spaenjers (Journal of Financial Economics, 2015) tracked the five First Growths from 1900 to 2012 and found a real return of about 4.1% a year net of storage and insurance, against 5.2% real for equities over the same 112 years. The flagship Investing in Fine Wine guide on this hub goes through that study and the smoothing that understates wine’s correlation with stocks; this guide takes the finding as read and asks a narrower question: how has Bordeaux done inside the fine-wine market since Liv-ex began publishing a tape in 2000?

The answer is: worse than the rest of it, for most of the years since 2011. The Liv-ex Bordeaux 500, the broadest Bordeaux index, lost 13.4% between May 2023 and the spring of 2024 (Liv-ex via wein.plus, May 2024), then fell 11.3% in calendar 2024 (Liv-ex, The Fine Wine Market in 2024) and 6.7% in 2025 (Liv-ex via The Drinks Business, December 2025). Over the five years to May 2026 it was down 16.5%, and its twelve-month reading in May 2026 was still fractionally negative at −0.4% (Liv-ex via WineNews and Decanter).

The First Growths did a little better at the turn: the Fine Wine 50 rose about 2.5% over the last four months of 2025, its first sustained gains since early 2023 (Liv-ex via Cult Wines and Vin-X), and was the best sub-index again at +0.7% over the four months to April 2026 (Decanter). That is a bottoming led by the most liquid names, which is how wine cycles usually turn.

Our own tape tells the same story at the benchmark level. Invest Alternative keeps month-end copies of the Liv-ex Fine Wine 100 and 1000 from July 2021 in its radar data store. On that series the Liv-ex 100 rose 6.9% in 2022, then fell 14.1% in 2023, 9.1% in 2024 and 2.5% in 2025; it peaked at 424.35 on September 30, 2022, bottomed at 309.27 on August 31, 2025, a 27.1% drawdown, and closed August 2026 at 322.92, 4.4% off the low and 1.1% higher for the year. Bordeaux is the largest component of the Liv-ex 100, so when it moves, Bordeaux is most of the reason.

Our Fine Wine sub-index, built on the same Liv-ex 100, read 103.27 on September 8, 2026, up 3.27% over twelve months and 0.66% over thirty days, at a 1.1% weight in the IA Composite. These are our copies of a published index and our own construction on it, not a market-wide measure of Bordeaux prices.

The gap between the marketing figure and the real one is wider in Bordeaux than anywhere else in wine, for three reasons. The indices are built from the wines that trade, which are the wines that held their value. The index is a mid-price, so it omits the spread and the 10% a merchant exchange charges a seller. And the return most private buyers actually earned is not the index return but the return from en primeur release to resale, which for most vintages since 2016 has been negative. The honest expectation for a case bought at the prices of September 2026 is a mid-single-digit gross return in a good decade, a loss in a bad one, and around two points a year given up to carry, commission and tax either way.

Bordeaux prices move in long waves driven by whoever the marginal buyer is, and knowing the last three waves is the best protection against being the marginal buyer at the top of the fourth. The first modern boom ran from the 2005 vintage, which Robert Parker called “the perfect vintage”, through the 2008 crisis, which the Liv-ex 1000 survived with a 10.5% fall while the S&P 500 lost 57% (Liv-ex). Wine looked defensive, and the money noticed.

The second wave was China. Hong Kong cut its wine duty from 40% to zero on budget day, February 27, 2008 (Hong Kong Inland Revenue Department), and mainland demand for one name in particular did the rest. In the five years to mid-2011 Lafite’s price rose more than twice as fast as its First Growth peers, and by 2010 Lafite traded at a 129% premium to the other four (Liv-ex, October 2021). The Lafite “eight” episode, the Sotheby’s Hong Kong sale and the Fine Wine 50’s June 2011 peak, the three scenes of the cold open, were the last months of that wave. What followed was slow and total. The index gave back 47.6% of its rise by January 2012 and 76.4% of it by 2014, and the Asian buyer base whose Bordeaux share was 95.5% in 2011 fell below 75% in 2017 and 50% in 2021 (Liv-ex).

Bordeaux went from about 90% of the secondary market in 2010 (Liv-ex via wein.plus) to 33.3% of volume and 36.2% of value in 2024 (Liv-ex), and nothing since, not the 2016 vintage, the 2019 campaign or the 2020–22 bull run, has restored that share.

The third wave was the one every scarce asset rode: near-zero rates and stimulus pushed the Liv-ex 1000 up 13.1% in 2022 on our tape and Champagne’s index up 93.9% between March 2020 and October 2022. Bordeaux rose less than Burgundy or Champagne on the way up: the Bordeaux 500 gained just 2.9% over the two years to the 2023 Liv-ex Classification, against 19% for the Liv-ex 1000 (Liv-ex via wein.plus, 2023). That was the one consolation of the way down. From the autumn 2022 peak to August 2025 the Liv-ex 100 lost 27.1% and the Liv-ex 1000 28.8% on our tape, the longest and deepest correction in Liv-ex’s 25-year history.

The turn came in the last four months of 2025, when Liv-ex bids reached £31M in December, the highest since April 2023, and the Fine Wine 50, 1000 and Bordeaux 500 all posted their strongest monthly gains in more than three years (Liv-ex via The Drinks Business and Vinetur, December 2025).

The pattern across all three cycles is that Bordeaux’s buyer changes each time: the London and New York collector in 2005–08, the Chinese trade buyer in 2009–11, and in 2020–22 a platform-driven cohort led by US collectors, who became the largest group on Liv-ex at 35% of purchases by value in 2024 (Liv-ex). After a 15% US tariff on EU wine took effect in August 2025, that share fell to under 17% by late 2025 (Liv-ex). Each new buyer arrives with the previous buyer’s price as an anchor and leaves with a loss. Buy Bordeaux when its share of trade is rising from a low and its marginal buyer has gone home, which describes 2014 and the end of 2025, and not any year in which a First Growth gains 20% on a label change.

Nobody buys Bordeaux from a château. The region has sold its wine through a three-tier chain since the 17th century, and the chain explains both the futures system and why a private buyer pays what they pay. The estate sells to négociants, the Bordeaux merchant houses that hold the wine and distribute it worldwide, but only through a courtier, a licensed broker who negotiates price and quantity and takes a 2% commission on every transaction (Wine Cellar Insider, FINE+RARE, iDealwine). The négociant sells to importers and merchants abroad, who sell to you.

A First Growth allocates its release across dozens of négociants, in tranches, and the négociant’s margin on the ex-château price gives the “ex-négociant” price the trade quotes: Lafite’s 2025 was released at €280 ex-château and €336 ex-négociant, a 20% step (Liv-ex closing report via Vinetur, June 2026).

The chain is where price discovery for a new vintage happens: the courtier advises the château what the market will bear, the château sets a price, and the négociants take their allocation or, as they increasingly have since 2022, take less of it and let the château hold stock. It is also why the same wine has two prices. The ex-négociant release is a wholesale number; a London or New York merchant’s offer adds a margin, and a Liv-ex bid is set by other members of the trade who all know what the release was. When Lafite 2023 traded at £4,180 after being offered to private buyers at £4,920, the £740 gap was the margin the chain had built into the release, marked back down by the market.

The secondary market runs through Liv-ex, a members-only exchange for the trade; a private investor buys and sells through a member merchant, who quotes against Liv-ex prices, which makes Liv-ex’s data the market’s tape. Bordeaux was the deepest and most liquid segment throughout the downturn, and in November 2025 it took 42% of traded value, the highest of any region (Liv-ex via Harpers and Trading Grapes). The lead is no longer automatic: in one month of mid-2026 Burgundy overtook Bordeaux as the most traded region by value, the first time it had done so since 2022 (Liv-ex via Harpers and The Drinks Business, August 2026). The auction houses, Sotheby’s, Christie’s, Acker and Zachys, are where old vintages and large formats clear, on a very different fee structure, as the buying section sets out.

On the other side of your trade, then, is a professional: a négociant unloading allocation, a merchant balancing stock, a platform rebalancing, or a collector selling through one of them. That is the difference from Burgundy or Napa, where the other side is often a single collector. Bordeaux’s liquidity is real because the counterparties have inventory to move, and its spread, mid-single digits on a First Growth in normal markets, widens sharply whenever the négociants are long and the private buyer is absent, which was the state of the market from mid-2023 to late 2025.

En primeur is the sale of a vintage while it is still in barrel, 18 to 24 months before it is bottled and delivered, and it is the mechanism that turned Bordeaux from a luxury into a financial asset. The sequence is fixed. In late March or early April the châteaux open their cellars for a week of barrel tastings by critics and trade buyers, and the scores that follow set the mood.

From late April into June the estates release prices, one at a time, each watching the reception of the last; the courtiers place the allocations with négociants, the négociants offer to merchants worldwide the same day, and merchants offer to private clients within hours, usually as a price per twelve-bottle case in bond. The buyer pays in full at once. The wine arrives roughly two years later, and until then the buyer owns a contract with a merchant, not a case.

The system exists because it finances the château, which is paid two years early and shifts the risk of holding stock to the négociants and, through them, to you. In exchange the buyer is supposed to get the lowest price the wine will ever be offered at and access to wine that will be scarce once bottled. The first has been false for most vintages since 2016. The second holds only for estates that release less than the market wants, and the First Growths, at roughly 100,000 to 300,000 bottles each, are not among them except in a campaign priced to sell.

Château Latour worked this out first. In a letter to négociants dated April 12, 2012, Frédéric Engerer, who runs Latour for François Pinault, said the 2011 would be the last vintage of either Latour or its second wine, Les Forts de Latour, sold en primeur; from then on the estate would release wine only when it judged the wine ready, seven years on for Les Forts and 10 to 12 for the grand vin (Decanter, Liv-ex, April 2012). Latour 2012 duly appeared in March 2020 through 30 négociants, the first new grand vin the estate had released in eight years (Decanter).

Latour’s exit was the clearest statement anyone has made that the château, not the futures buyer, should capture the appreciation between barrel and maturity. No other First Growth has followed, which tells you how much the others value the financing.

For a private investor a futures purchase is a two-year zero-coupon bond with a wine attached: you give up the use of your money, take the merchant’s counterparty risk for two years, and get a case at the release price. It pays only if the release is below what the same wine, or a comparable back vintage, can be bought for in bottle at the same moment. That single comparison is the whole discipline of en primeur, and the next section shows how rarely the châteaux have let it come out in your favour.

The campaigns since 2019 are a controlled experiment in what happens when a release price is set with and without regard to the secondary market. The 2019s, released in June 2020 into the first pandemic summer, came out on average 21.3% below the market price of the 2018s, 17.6% below the 2016s and 35.6% below the 2010s (Liv-ex, 2020). Merchants sold out of the best names within hours, and by the end of July trade in the new vintage was running 400% ahead of the 2018s a year earlier (Decanter, The Drinks Business, Cult Wines). It is the one campaign since 2016 that made money for the people who bought it, and the reason is not mysterious: the prices were below the back vintages.

The estates read 2019 as a sign that demand had returned rather than that price had been right. The 2020s and 2021s went out at prices this guide did not re-verify beyond Lafite’s 2021 release at €470 a bottle ex-négociant, into a bull market that was over by mid-2022.

Then came the 2022s. Liv-ex members polled in May 2023 expected releases to rise 7.7% on the 2021s; the average increase was 20.8% (Liv-ex, Decanter). Lafite 2022 came out at €580 ex-négociant, up 23.4%, or £7,140 a case in London (Liv-ex, June 2023). Decanter called the campaign “another missed opportunity”. The wines were excellent; they have traded below release ever since, and the 2022s are a large part of why the Bordeaux 500 fell further in 2025, at 6.7%, than any other Liv-ex regional index (Liv-ex via WineNews, January 2026).

The 2023 campaign was the first to admit the error. Lafite released at €396 ex-négociant, a 31.7% cut, offered by UK merchants at £4,920 a case; Léoville Las Cases cut 40% (Decanter, Liv-ex, May 2024). The reception was warm and the result was still a failure, because the cuts brought the 2023s level with the back vintages rather than below them, and Lafite 2023 traded on Liv-ex at £4,180, 15% below its UK release price, within weeks of the offer (Liv-ex, “Bordeaux 2023 finds the secondary market,” May 2024).

A Wine Lister survey of 50 trade heads before the 2024 campaign called for an average cut of 31% on the 2023 prices (Decanter, April 2025), and the 2024s delivered it: Haut-Brion at €240, down 23.8% and equal to its 2014 release; Ausone at €312, its lowest since 2013; Cheval Blanc at €276, down nearly 30% and below any release back to 2008 (Liv-ex via Decanter); and Lafite at €288 ex-négociant, which Decanter noted made the new vintage the cheapest Lafite on the market (Decanter, EHL Insights, Wine-Intelligence, May 2025). Even so, most wines were released above the fair value Liv-ex’s models implied, some by 40 to 60% (EHL Insights), and the campaign’s value fell about 60% against the 2023s.

The 2025 campaign, run in May and June 2026 for a vintage critics placed among the stronger of the decade, tested whether the reset had worked. Liv-ex’s closing report called it “middling”: selective buying, cautious trade participation, strong competition from older vintages, and demand for only a handful of wines. The successes were Cheval Blanc, Margaux and Lafite, each released at €280 ex-château (Lafite’s ex-négociant price was €336), and Batailley among the value names (Liv-ex via Vinetur and The Drinks Business, June 30 and July 1, 2026).

The €336 is 16.7% above the 2024’s €288. Lafite raised its price into a market that had rejected the last three campaigns, and got away with it only because the 2024 had been priced so low that the 2025 was still cheaper than its 2019, 2020 and 2022 in bottle. Liv-ex’s line was that the 2025 is “an opportunity rather than a mistake” (The Drinks Business, June 2026), fair and also faint praise.

The 2024 and 2025 campaigns failed at lower prices for three reasons, none of them about the wine. The back vintages were cheaper: after three years of falling prices a buyer could have the 2019 or 2016 in bottle, with a bottle score, for the same money or less. The buyer base had thinned: the US, Liv-ex’s largest buyer in 2024, faced a 15% tariff from August 2025 and its share fell to under 17%, and the Asian buyer had not returned at 2011 scale. And the négociants were full: the 2022s and 2023s they had taken on allocation sat on their books below cost. A campaign succeeds only when the release is below the market and the chain has room to hold it; neither held in 2025.

For most releases since 2016, a buyer would have done as well or better buying the bottled wine, with 2019 the exception; that is Liv-ex’s own finding, repeated in its en primeur reports since 2019, and the comparison behind it is the single most valuable habit a Bordeaux buyer can acquire. The reason is structural. A release price is set by the château at the top of the range the négociants will accept; a back vintage is priced by a secondary market in which the seller competes with every other case on offer.

Put numbers on it with Lafite. The 2023, released at £4,920 a case in bond, traded at £4,180 within weeks: a buyer at release was 15% behind a buyer in bottle before storage or commission. The 2022, released at £7,140, has traded below that ever since; the exact discount as of September 2026 was not verifiable for this guide, but the later releases at €396 and €288 make it certain. The 2024, at €288, was released below every Lafite then on the market, which is why it sold and why its buyer is, for once, ahead. The futures buyer wins only when the château prices below its own back catalogue, roughly one campaign in four, and loses in the other three.

There is one more reason to prefer bottle to barrel, and it is about information. A barrel sample tasted in March is not the wine; it is a blend assembled for the tasting week, which will spend another year in oak, be fined, be bottled and then evolve. Scores from barrel are provisional and are routinely revised on bottle, in both directions. A back vintage has a bottle score, a track record of trades, and a drinking window that critics have actually observed. You are paying for a wine rather than a forecast, and the price difference between the two is usually in your favour.

Bordeaux is an agricultural product, so the same château’s 2016 and 2017 are different assets, and the vintage is the second thing after the name that sets the price. The trade’s working ranking of the last 25 years on the Left Bank, from the Wine Cellar Insider’s vintage chart, runs 2016, 2022, 2020, 2010, 2019, 2023, 2018, 2009, 2015, 2005 and 2000; on the Right Bank, where Merlot ripens differently, 2022, 2009, 2018, 2015, 2016, 2019, 2023, 2010, 2005 and 2000. Parker called 2005 “the perfect vintage”; 2009 and 2010 were each proclaimed vintages of the century, and 2016, 2018, 2019 and 2020 have all been nominated since. The lighter years, 2007, 2011, 2013, 2017, 2021 and 2024, are wines to drink rather than hold; they can be bought cheaply and they stay cheap.

Two archetypes matter for holding periods. A classic, structured year such as 2010 or 2016, with high tannin and acidity, takes 20 to 30 years to reach its plateau and holds it for a decade or more; those are the long investments, because you own the whole approach to maturity. A ripe, opulent year such as 2009 or 2018 drinks well earlier and can peak sooner, which shortens the runway and, historically, the appreciation. A First Growth from a classic year bought at 10 years old is entering the period when it is both scarce enough and drinkable enough to command a premium; bought at 30 it has most of its appreciation behind it and its provenance risk in front.

Scores are the third lever, and the authority behind them has fragmented. Robert Parker built the 100-point scale on the 1982 vintage, which he called great when his peers called it too ripe, and for 30 years a Parker point was worth real money to a château.

He handed Bordeaux en primeur to Neal Martin in February 2015 and retired in May 2019; Michelin took 40% of The Wine Advocate in 2017 and full ownership in 2019, and since then the publication has scored as a team under editor-in-chief Joe Czerwinski (Decanter, Wine-Searcher, The Wine Advocate). Martin writes for Vinous with Antonio Galloni; Jane Anson, Jancis Robinson on a 20-point scale, and James Suckling complete the set that moves prices. In practice a 95-plus consensus score is the floor for investment grade, and a 100 from any of the major critics can move a wine’s price by a multiple, but the era in which a single re-score could re-price a vintage ended with Parker, and an investor should weight the average of the critics over any one of them.

Liquidity, not scarcity, is what an outsider should buy, and in Bordeaux the two are inversely related. The First Growths are the most liquid wines in the world: bid and offered every day on Liv-ex, in every strong vintage back to 1982, in a spread that runs mid-single digits in normal conditions. They also led the market off its August 2025 low, as the record section shows. Inside the five, Lafite’s 129% premium to its peers in 2010 compressed for a decade and, per Liv-ex, began to widen again from 2021; Latour, on its own release schedule, has had the fewest failed releases; Margaux and Mouton sit between; Haut-Brion, the smallest and least known in Asia, is habitually the cheapest of the five for the same score, a discount or a warning depending on your horizon.

The super-seconds, the seven names listed in the first section, behave like mid-caps: they trade actively in strong vintages at a fraction of First Growth prices, on spreads roughly twice as wide. Below them liquidity falls off a cliff; a fourth growth may take weeks to sell at the price the index suggests, and the list’s lower tiers are collectors’ wines, not investments.

On the Right Bank the bid is concentrated in about eight names: Pétrus and Le Pin in Pomerol, whose production (roughly 30,000 bottles for Pétrus, a figure this guide did not re-verify; 400 to 600 cases for Le Pin, per Sotheby’s) makes them trophy assets that clear at auction as often as on the exchange; Cheval Blanc and Ausone in Saint-Émilion, where the 2022 classification exit made no difference to price; and Angélus, Pavie, Figeac and Lafleur behind them. Cheval Blanc’s 2025 was one of the three clear successes of the 2026 campaign at €280 ex-château, which says the market will still pay a First Growth price on the Right Bank for a First Growth name.

The second wines, Carruades de Lafite, Les Forts de Latour, Pavillon Rouge, Le Petit Mouton and Clarence de Haut-Brion, trade on the exchange and rose furthest in the China years on the strength of the label; Carruades’ 2024 release at €120 was a 23.2% cut. They are liquid, cheap, and the first thing a fading buyer base drops. On formats, the trade’s unit is the original wooden case of twelve, and broken cases and loose bottles trade at a discount; magnums and larger formats carry an auction premium because they age more slowly and are rarer, but are less liquid on the exchange, so hold them only if auction is the exit you plan for.

Every route into Bordeaux ends at a member of the trade, and the fee depends on which member and which direction. The cheapest way to own a case is through a merchant, in bond. Berry Bros. & Rudd, Farr Vintners, Lay & Wheeler and Justerini & Brooks in London, and Zachys and Sokolin in the US, sell cases from stock or on allocation at a price that includes their margin; there is no separate commission on the way in. On the way out the merchant’s exchange charges the seller: BBR’s BBX takes about 10% of the sale price (BBR published terms), on top of a Liv-ex spread that runs mid-single digits on a First Growth in a normal market and wider in a stressed one.

Auction fees are built for the seller of a cellar, not the buyer of a case. Sotheby’s charges a flat 24% buyer’s premium on wine and spirits, the one category it excluded when it raised its general premium to 28% from February 13, 2026 (Antiques Trade Gazette, The Value), plus a seller’s commission of about 10%, negotiable or waived for large consignments. Christie’s raised its buyer’s premium to 27% on lots up to $1.5M in September 2025 (as reported by Wine Spectator and The Art Newspaper) and left wine out of its September 1, 2026 rise to 28% (ARTnews). Acker and Zachys, the two American specialists, each publish a 25% buyer’s premium (house FAQs and conditions of sale, September 2026), so the round trip there is wider than Sotheby’s before any seller’s commission is counted.

A buyer bids net of the premium, so the seller of a single case at auction receives about 70 to 75 cents of what the buyer pays. Auction earns its fee for mature wines, large formats and full cellars; for a recent First Growth in bond, the exchange is the better exit by a wide margin.

Storage is the smallest line and the most important. A bonded warehouse holds wine with duty and VAT deferred, in professional conditions, and attaches a rotation number to every case that follows it from owner to owner; that record is what lets the case sell into Liv-ex pricing later. London City Bond charges £14.40 a case a year via Farr Vintners (from June 2025) and Octavian £16.57 above a six-case minimum, both plus 20% VAT with insurance bundled (published tariffs). Call it £17 to £20 a case a year, under 0.5% of a First Growth case. A US buyer can hold in a UK bond indefinitely, which keeps the wine in the deepest market and defers duty and the 15% tariff in force since August 2025; shipped to the US, the wine loses its in-bond status and much of its resale liquidity.

Managed platforms, Cult Wines from $10,000 at 2.95% a year and Vinovest at 2.85% tiering to 2.25%, from an entry minimum reported at $5,000 since March 2026 (published fee schedules, September 2026), buy, store and sell for you, and the flagship guide’s arithmetic shows the fee costing more than the tax over seven years. Bordeaux is where they add least, because a First Growth is the one wine a merchant will sell you at a checkable price without a relationship. Pay a platform for allocation you cannot otherwise reach; do not pay it for Lafite. The sister guide, Investing in Wine Through Platforms and Funds, ranks the platforms and funds on their published fees and minimums.

The United States treats wine as a collectible, and the treatment is unambiguous. IRC §1(h)(4) and (5) point to §408(m)(2), whose list of collectibles includes “any alcoholic beverage”, so a gain on a case held more than a year is taxed at a maximum federal rate of 28% rather than the 15–20% that applies to stocks (IRS Topic 409). The 3.8% net investment income tax under §1411 applies above $200,000 of modified adjusted gross income for a single filer and $250,000 for a couple, and state tax comes on top. A case held a year or less is ordinary income. The 28% is a ceiling: a taxpayer in a lower ordinary bracket pays the lower rate.

Basis includes what you paid to acquire the case, so a buyer’s premium or a merchant’s margin embedded in the invoice reduces the eventual gain, which is one more reason the invoices belong in the provenance file. The 2025 tax legislation left all of this unchanged, and the flagship guide walks through the UK’s partial exemption, where HMRC’s wasting-asset rule shields cheap wine but explicitly not the First Growths a US investor would hold. None of this is advice; it is the terrain, and the worked example below applies it.

The example uses one twelve-bottle case of Château Lafite Rothschild 2023 in bond, bought from a merchant at $5,500, which is about £4,230 at $1.30 to the pound and close to the £4,180 at which the wine traded on Liv-ex after release. It is held seven years, stored at London City Bond, sold through a merchant exchange at a 10% commission, and taxed at the 28% federal rate with no state tax or NIIT. Three scenarios, all Invest Alternative arithmetic.

In the first, the case compounds at 5.3% a year, the Liv-ex 100’s compound rate from its 2004 base of 100 to the 322.92 on our tape at August 31, 2026, before costs, and a generous assumption for a single vintage of a single château. After seven years it is worth $7,895. Storage at £14.40 plus VAT is £17.28 a year, about $22.50, or $157 over the hold. Selling at 10% commission clears $7,105. The taxable gain is $7,105 less the $5,500 basis, or $1,605, and the federal tax on it is $449. You keep $7,105 less $449 less $157: $6,499, a gain of $999 on $5,500, or 2.4% a year. The gross return was 5.3%; the exchange took 1.6 points a year, the government 0.9, the warehouse 0.4.

In the second, the case follows the Bordeaux 500’s actual five-year pace to May 2026, a 16.5% fall, or about −3.5% a year, for the whole hold. It is worth $4,286 in year seven, $3,858 after commission, and $3,701 after storage: a loss of $1,799, or 32.7% of the money. The capital loss of $1,642 is deductible against other capital gains under §165(c)(2), because the case was held for investment; the storage is not.

In the third, the buyer took the en primeur offer at £4,920, about $6,400, and the wine follows the first scenario. The case is still worth $7,895 in year seven and still nets $7,105, but the basis is $6,400, so the gain is $705, the tax $197, and the buyer keeps $6,751 on $6,400: 0.8% a year. The futures buyer and the back-vintage buyer own the identical case; the difference between 2.4% and 0.8% a year is the release price.

Two things the example leaves out cut both ways. Currency is one: the case is priced in sterling, and a 10% move in the pound over seven years is larger than the storage and comparable to the tax. The exit price is the other: the example sells at the mid-price, and a seller in a stressed market may cross a spread of 5% or more on top of the commission, which would take the base case below 2% a year. What the example shows is the shape of every Bordeaux return: the gross is set by the market, and roughly half of it is spent on the way out.

Bordeaux’s liquidity is also its exposure, because the wines that are easiest to sell are the ones most worth faking. The largest fraud in the modern market was Rudy Kurniawan’s: two Acker auctions in 2006 sold about $35M of old Burgundy and Bordeaux, the FBI found some 18,000 fake labels in his California house in 2012, and in August 2014 he was sentenced to ten years and ordered to pay $28.4M in restitution and forfeit $20M (U.S. Department of Justice, SDNY). The oldest is a Bordeaux: the 1787 “Th.J.” Lafite that Christie’s sold for $156,450 in December 1985, whose engraving Bill Koch’s investigators concluded was made with a power tool, and whose source, Hardy Rodenstock, died in 2018 without producing the cellar (Christie’s, Decanter, World of Fine Wine). Both are set out in the flagship guide.

The specifically Bordeaux lesson is scale: Kurniawan faked Burgundy by the bottle, but Lafite was faked in China by the container. An adviser to the French government on wine exports said in 2012 that a Chinese buyer of Lafite had a one-in-two chance of a fake, analysts’ estimates ran as high as 70% (Underground Wine Letter, Recorded Future), and some 10,000 suspected fake Domaines Barons de Rothschild bottles were found in a house in Wenzhou (Decanter). None of those figures was measured; this guide repeats them as the trade’s estimates.

The château’s response set the regional standard. From February 2012 every bottle of Lafite and Carruades leaving the estate has carried a Prooftag bubble seal, a capsule with a 13-digit code and a random bubble pattern that cannot be reproduced, applied to the grand vin from the 2009 vintage and to Carruades from the 2010 (Decanter, Goedhuis Waddesdon, DBR Lafite); Margaux uses the same system. The consequence for a buyer is a line in the sand: Lafite from 2009 onward, still sealed, can be authenticated by anyone with the code; anything older is authenticated by its paperwork or not at all.

Provenance, in Bordeaux, means an unbroken in-bond record. A case that has sat in a UK bond since the négociant delivered it, with a rotation number that followed it through every change of owner, is what Liv-ex’s Standard-In-Bond contract defines as deliverable and what a merchant will buy without inspection. A case that left bond and came back needs a condition report and sells at a discount; a case with a gap in its history is a coin flip, and the odds worsen with age and rarity. Recent, in-bond, sealed and documented is the safest and most liquid form of the asset, and the cheapest per point of critic score as well.

The sequence for an outsider with $10,000 to $50,000 to put into Bordeaux is short and its order matters.

1. Open an account with a merchant that is a Liv-ex member and offers in-bond storage and a resale desk; in London that is BBR, Farr Vintners, Justerini & Brooks or Lay & Wheeler, and each will take a US client. Ask for the storage tariff, the seller’s commission and the insurance basis in writing before you buy anything.
2. Learn to read Liv-ex prices before you read merchant offers. A merchant’s quote for a First Growth should sit within a few percent of the Liv-ex market price; if you cannot see the print, you cannot judge the offer.
3. Buy back vintages in bottle, not futures, until a campaign passes the 10% test in the en primeur section. Start with one or two intact original cases of a First Growth or a super-second from a top-ten vintage that is between five and fifteen years old.
4. Leave the wine in bond. Do not ship it to the US, where it loses its in-bond status, incurs duty and the 15% tariff in force since August 2025, and leaves the market it trades in.
5. Keep a provenance file: the invoice, the bond’s receipt and rotation number, the merchant’s condition statement, and every subsequent storage invoice. This is the document that sets your tax basis and the document that sells the case.
6. Size the sleeve for a 27% drawdown and a seven-year hold: the flagship guide’s 1–5% of a portfolio is the right range, and the worked example shows why more does not help.
7. Plan the exit at purchase. A First Growth clears through the merchant exchange in days at a 10% commission; that is the route, and a case you would only be able to sell at auction is a case you should not buy as an investment.

The readings that would change the view, each with its level as of the dates given.

• The Bordeaux 500’s twelve-month change. Still −0.4% in May 2026 (Liv-ex). The first positive twelve-month print is the signal that the deepest market has turned; two consecutive negative prints below −5% would say the 2025 floor did not hold.
• The Liv-ex Fine Wine 100 on our tape. 322.92 on August 31, 2026, against a 309.27 low (August 31, 2025) and a 424.35 peak (September 30, 2022). A month-end below 309 reopens the drawdown; a close above 370, the October 2021 level, means the easy recovery is done.
• The 2026 en primeur campaign, spring 2027. Lafite, Margaux and Cheval Blanc released the 2025s at €280 ex-château. Releases at or below €280 for a comparable score would say the châteaux have accepted the market; a step back toward the 2022’s €580 ex-négociant would restart the cycle that failed in 2023, 2024 and 2025.
• Lafite 2022 against its release. The 2022 came out at £7,140 a case in June 2023. The day it trades at release again is the day the last bull-market vintage has been absorbed; until then the négociants are still long.
• Bordeaux’s share of Liv-ex trade. 36.2% of value for 2024, 42% in November 2025; by August 2026 Burgundy had taken the top spot in a single month for the first time since 2022 (Liv-ex via Harpers). Sustained months above 40% mean the liquidity buyer is back; a slide below 30% means the flows have gone to Burgundy, Italy and Champagne again.
• The US share of Liv-ex purchases and the tariff. 35% in 2024, under 17% in late 2025 after the 15% tariff took effect in August 2025. Removal of the tariff, or a US share back above 25%, is the single most bullish development available to Bordeaux.
• Liv-ex bid volumes. £31M in December 2025, the highest since April 2023. A run of months back below the 2024 average would mark that as a year-end flurry rather than a turn.
• Our Fine Wine sub-index. 103.27 on September 8, 2026, +3.27% over twelve months. A negative twelve-month print puts the floor back in question.

The one that matters most is the first. Bordeaux has been the laggard of the fine-wine market since 2011, and the case for owning it is that the laggard, at a five-year low in May 2026 with bids rising from December 2025 and a chastened en primeur system, is the cheapest liquidity in wine. That case holds while the Bordeaux 500 is flat to rising; it fails if the index resumes falling while Italy and Champagne recover, because then the buyer has not left the market, only Bordeaux.