Wine futures: how buying Bordeaux en primeur works
Updated
With wine futures you pay in full, wait two to three years for the bottles, and find out afterwards what you actually bought.
Wine futures, sold in Bordeaux as en primeur, are wine bought while it is still in barrel. Berry Bros. & Rudd says it in one line: "Buying En Primeur means you are buying a wine before it has been bottled." Barrel samples are tasted in April, prices come out over the following weeks, and the bottles reach a warehouse two to three years after the grapes came in. A UK buyer pays an in bond price that covers shipping but excludes duty and VAT, and both fall due only when the wine leaves the bonded warehouse. Nothing in the process is a contract for a named bottle on a fixed date. What you hold until delivery is a claim on a merchant. The reason to buy is allocation and price. The reason most collectors now hesitate is that Liv-ex's release-price series puts every Bordeaux vintage from 2016 to 2023 below its release price today.
What are wine futures, and what do you actually own?
A claim on your merchant for wine that does not exist in bottle yet. Not a case with your name on it in a warehouse.
The Wine Cellar Insider states the terminology plainly: "Buying wine En Primeur or purchasing wine futures are the same thing. When you buy wine En Primeur or as a future, the wines are still in the barrel." That is the line that separates futures from pre-arrival stock, where "the wine has been bottled, but has not shipped yet". Futures are earlier and looser. You commit on a barrel score, and the wine will not be bottled for many months after the tasting.
The delivery date is an estimate, and merchants say so. Berry Bros. & Rudd: "Exactly when the wines are shipped is down to when the producer thinks they're ready, so we can't guarantee exact dates." Their en primeur terms put delivery at "two to three years after the harvest", add that shipping dates "are only an indication and cannot be guaranteed with certainty", and reserve the right to scale orders back: "Some customers may not receive an allocation of their order or may receive less than their order amount." Ownership, in the same terms, "will pass to the customer upon settlement of all outstanding accounts."
So the counterparty matters as much as the château. You are extending an unsecured, interest-free, two-year loan to a wine merchant, and the only thing you get for it is a place in a queue.
Who takes a margin between the château and you?
Three of them, and in the 2025 campaign they added around 40% to the château's price before a UK private buyer saw a number.
Most Bordeaux châteaux do not sell direct. They sell through La Place de Bordeaux, a distribution layer that Insider Tasting describes as "7 900 growers, 300 Negociants and 95 Brokers in Bordeaux". The brokers, or courtiers, sit between château and négociant and "never take title to the wines (i.e. own them) but take a 2% margin on each transaction". Club Oenologique, which calls the courtiers "somewhat shadowy brokers", puts the négociants' reach at "two-thirds of all wine produced in Bordeaux, to merchants in some 170 countries", and dates the first attempt to license courtiers to Louis IX in 1243.
Liv-ex prices each rung. For the 2025 vintage, Liv-ex reports from its members' data an average ex-château price of €111.60 a bottle, an ex-négoce price of €134.00, a UK merchant margin of 14.5%, and an ex-London release price of €156.70. That is the anatomy of the number in your offer email.
The system persists because the cash flows the right way for the château. Jean-Christophe Mau, of the Bordeaux négociant house Yvon Mau, gave the asymmetry its sharpest form in a line quoted by 67 Pall Mall: "Négociants can refuse to buy, but châteaux can't turn off the tap." Classified growths and their equivalents are, on Insider Tasting's figures, "approximately 5% of volume of Bordeaux production but 20% of its value", which is why a few dozen estates dominate a campaign that nominally covers a region. Use the Bordeaux producer atlas to see which of them released in a given year and where they sit relative to each other.
When does a campaign actually run?
It starts with one week in April and finishes at a warehouse door about two years later.
The Union des Grands Crus de Bordeaux sets the opening date. Its notice reads: "From 20 to 23 April 2026, the Union des Grands Crus de Bordeaux welcomes you to Bordeaux for "En Primeurs" Week", with 132 member châteaux and Monday's tasting at Hangar 14 on the Quai des Chartrons. Trade and press taste, critics publish, and only then do prices appear.
Releases come in tranches. Savor the Harvest describes the pattern: "Depending on how the first tranche sells, châteaux may release a second larger tranche and possibly a third. The price may increase with each tranche, but not always, it depends on each individual château." At each release the merchant chain gets a choice rather than an instruction: "They can accept, ask for a discount or decline to purchase their allocation." In a weak year that choice is exercised, and the campaign stretches out. Robert Mathias, buying director at Lay & Wheeler, described the 2025 release pace as "slow and uneven".
Bordeaux is the loudest en primeur market, not the only one. Private Cellar lists when each region sells forward:
| Region | When it is released en primeur |
|---|---|
| Bordeaux | "The spring following the harvest" |
| Burgundy | "Approximately 15 months after the harvest by most growers" |
| Rhône | "12-18 months after their harvest" |
| Port | In declared vintages, "usually released 18 months following the harvest" |
| Italy | Varies by region; top Piemontese wines and Brunello di Montalcino "up to four years after harvest" |
| California | Joseph Phelps, including Insignia, "usually offered in November each year" |
The further from harvest a region sells, the less of the guesswork you are carrying, and the smaller the discount you should expect for carrying it.
What do you pay now, and what do you pay later?
You pay the in bond price now. Duty, VAT and storage arrive later, and they are not small.
Berry Bros. & Rudd's guide separates the two: the quoted price "includes the cost of shipping to our UK warehouses and, subsequently, delivery to you", and "You don't need to pay Duty and VAT until the wines are withdrawn from Bond." Private Cellar adds the running cost, since wine left in bond means "annual storage is charged which also covers insurance".
HMRC lists a rate of £30.62 per litre of pure alcohol for wine between 8.5% and 22% ABV, on the rates page updated 1 February 2026. Work that through a standard case and the number stops being abstract: a 75cl bottle at 13.5% ABV contains 0.10125 litres of pure alcohol, so duty is about £3.10 a bottle and roughly £37 on a twelve-bottle case. Raise the alcohol and the bill rises with it, because the charge follows the alcohol rather than the price.
VAT then sits on top of that total. HMRC states both the timing and the base in its tax warehousing notice: "VAT is due on the last supply of goods in the warehouse regime. It must be accounted for and paid (or deferred) when the goods are removed from the warehouse regime to home use (this is usually the point at which duty becomes payable on the goods)", and "You must include any excise duty in the value." Two consequences follow. Wine held and traded in bond stays outside the charge. Wine you drink does not, and the taxable value is the last in bond supply price, not necessarily the release price you paid in 2026.
US buyers have a further line. The United States was set to impose a 15% tariff on EU wine and spirits from 1 August 2025, and the European Commission confirmed in the run-up that "while some goods, like aircraft and cork, have secured exemptions, wine and spirits are not currently included." That was the position in mid-2025. A futures order placed today clears customs under whatever regime exists on the day it lands, so check the rate that applies then, not the one that applied when you paid.
Put every line in one place before you commit with the landed-cost calculator.
Have wine futures actually made money?
Rarely, over the last decade. The En Primeur 2025 closing report by Sophia Gilmour, market analyst at Liv-ex, tracks release prices against current market prices for "38 of the 50 current Bordeaux 500 components for which we have a full set of release prices between 2008 and 2025", and the record is unflattering.
| Vintage | Ex-château (€/btl) | Ex-London release (€/btl) | Current market price (€/btl) | Current return |
|---|---|---|---|---|
| 2008 | 59.1 | 87.1 | 152.6 | +75.2% |
| 2009 | 177.8 | 269.9 | 236.4 | -12.4% |
| 2010 | 207.1 | 279.4 | 232.5 | -16.8% |
| 2011 | 112.1 | 145.8 | 137.4 | -5.7% |
| 2012 | 90.8 | 120.9 | 153.2 | +26.6% |
| 2013 | 82.4 | 108.7 | 128.4 | +18.2% |
| 2014 | 92.6 | 124.6 | 136.9 | +9.9% |
| 2015 | 136.2 | 184.4 | 189.8 | +2.9% |
| 2016 | 155.9 | 218.2 | 202.3 | -7.3% |
| 2017 | 133.2 | 183.8 | 131.6 | -28.4% |
| 2018 | 156.1 | 219.0 | 158.5 | -27.6% |
| 2019 | 121.2 | 171.3 | 150.6 | -12.1% |
| 2020 | 153.7 | 215.8 | 150.4 | -30.3% |
| 2021 | 150.1 | 212.3 | 120.5 | -43.3% |
| 2022 | 184.9 | 259.5 | 185.7 | -28.4% |
| 2023 | 132.9 | 189.7 | 153.8 | -18.9% |
| 2024 | 99.4 | 139.6 | Not yet physical | Not yet physical |
| 2025 | 111.6 | 156.7 | Not yet physical | Not yet physical |
Five of the sixteen settled vintages show a gain. Eight consecutive vintages, 2016 through 2023, are underwater. The 2021 vintage is the worst of them at -43.3%, and Liv-ex lists 2009, 2010, 2021 and 2022 as "the worst offenders" for over-ambitious release pricing. The 2015s have returned 2.9% since release, which after ten years of storage is a loss in real terms.
Liv-ex reports a second return column that most buyers never look at: the return at the moment the wine became physical, before any long hold. On that measure only four of the sixteen vintages were ahead by delivery day. Liv-ex is explicit that its comparison "does not take into account cost of carry or the risk of buying En Primeur", so the real figures for a private buyer are worse than the table by the amount of storage, insurance and foregone interest over two to twelve years. The market index shows what the same wines have done as a group once they are trading in bottle.
Why did the 2024 and 2025 campaigns go so badly?
Because release prices stayed above the price of better-regarded vintages already sitting in bond. Liv-ex says exactly that of 2024: it was "unsuccessful, priced higher, on average, than other better-regarded vintages already on the market."
Its account of what happened next describes a distribution system stalling. It was "not just private clients, but merchants and negociants that found themselves either unable or unwilling to take on stock of the (poorly rated but sometimes aptly priced) 2024s", and "Without sufficient demand from the rightly-dubious collector, there was neither warehouse space nor cash available to buy what could not be sold." The 2024 average ex-château price of €99.40 was the lowest in the series since 2014. It still did not clear.
2025 had the opposite quality problem and the same commercial one. Gavin Quinney, whose crop report Liv-ex publishes, called it "an excellent vintage for Bordeaux but it's another dastardly small one, unfortunately, with low yields". The closing report found "The 2025s received very solid scores, generally ranking amongst the best of the past decade. They were released at similar levels to the current pricing of the 2019s and 2020s, which frequently came out on top as the clear alternative." Ex-château pricing rose about 12% on 2024. "The majority of merchants we spoke to thought release prices were too high to overcome this year's hurdles", and others cited "the breach of trust between chateaux and collectors, and the latter party's subsequent disengagement".
Marc Ditcham, fine wine buyer at Corney & Barrow, said many châteaux appeared to "ignore repeated warnings from négociants, British merchants and consumers about a difficult global economic backdrop". Max Lalondrelle, managing director of fine wine purchasing at Berry Bros. & Rudd, reported the campaign "worked" for his firm, which offered "around 60 wines this year", roughly half the number of a decade ago. Liv-ex found demand had narrowed to a small pool of successes, with "Cheval Blanc, Margaux, Lafite, Batailley coming up almost ubiquitously", and described a campaign in which "demand was more selective than it may have been in the past".
The lesson generalises past Bordeaux. A forward price only works when the spot market for comparable vintages is higher, and for most of the last decade it has not been.
What can go wrong that has nothing to do with price?
Your merchant can fail while holding your money and no wine.
Premier Cru, a Berkeley retailer, is the case every futures buyer should know. Its owner John Fox sold pre-arrival wine and futures the business did not have, backed by what court documents call false purchase orders for non-existent "phantom wine". Around 4,500 customers were left with losses that prosecutors put at $45m to $50m, and Fox was sentenced on 14 December 2016 to six and a half years in federal prison. The Wine Cellar Insider says the same collapse is its reason to check a merchant first: "The recent collapse of Premier Cru, or the previous debacle of the Rare LLC Company and Carolina Wines makes it imperative that the store you buy from has been in business, with a track record of selling and delivering futures."
Three smaller failure modes are more common and less dramatic:
- Short allocation. Merchants scale orders down when a wine oversells, and their terms permit it.
- Slipping delivery. The bottling date belongs to the producer, so a two-year wait can become three.
- The bottle is not the sample. You commit on a barrel score, and the wine is bottled long after the tasting.
None of these is a reason to avoid futures. All of them are reasons to spread orders across merchants, keep the paperwork, and treat a large single-merchant exposure as a credit decision rather than a wine decision.
Should you buy wine futures at all?
Only when the forward price beats the spot price for a comparable bottle. Liv-ex frames the test as one question: "can I buy another similarly or better rated vintage already on the market at a lower price?"
Three conditions make a futures purchase defensible. The wine is scarce at release and gets scarcer, which is true of a short list of estates and no more. The release price sits below the current market price of an equal or better vintage in bottle, after you have added duty, VAT, carriage and years of storage. And the format you want, magnums, large formats, original wooden cases, is hard to source later at any price. Miss all three and the patient move is to buy the physical vintage that is already trading, where the wine, the score and the condition are known quantities.
Liv-ex ends its 2025 report on the case for patience: "In short, this was not a knock-out campaign, but neither have those of the past that we now look fondly back on been either." That is the honest position. Futures are not a broken idea, they are a badly priced one, and pricing changes.
If the bottle in question is already yours, the question shifts from what to pay to when to open it, which is what drink-now is for. If you are weighing a futures allocation against a bid on a mature case of Pétrus, price the auction route properly first with what a £1,000 hammer actually costs.
See the release price next to what the bottle actually sold for
The release price is a claim about the future. The auction record is what buyers paid. Every wine page here carries the second one, so you can hold a 2026 offer against the realised prices of the same estate's earlier vintages before you commit two years of capital.
Join the newsletter and we will send the campaign reads that matter: which releases undercut the physical vintages already in bond, and which ones ask you to pay more for less.