Investing in Champagne: What the Price Data Says
Updated
Growers and houses in Champagne sit on one committee that decides every summer how much the appellation may sell, and it has cut that number four years running.
Investing in Champagne means buying prestige cuvées with a working secondary market, holding them in bond, and selling to someone who wants that bottle with better provenance than a shop can give them. The record is better than the rest of fine wine and thinner than the sales pitch. Liv-ex reports the Champagne 50, which tracks "the most recent physical vintages of the 16 most actively traded champagnes", at 500.7 in August 2026: up 2.6% over a year, down 8.4% over two years, up 8.7% over five. That five-year figure is the strongest of any Liv-ex regional index, against Bordeaux 500 at minus 18.3%. The supply story is the reason to look at all. On 23 July 2026 the Comité Champagne set the marketable yield at 8,800 kg per hectare, around 250 million bottles, the fourth consecutive cut and, according to Refine Drinks, the lowest in the modern era outside the Covid year.
Has investing in Champagne actually paid?
Over five years, better than anywhere else Liv-ex measures. Over two years, it lost money like everything else. Both facts come off the same board, and which one you are shown tells you who is showing it.
| Liv-ex index, August 2026 | 1 year | 2 years | 5 years |
|---|---|---|---|
| Champagne 50 | +2.6% | -8.4% | +8.7% |
| Italy 100 | +2.9% | -4.2% | +4.6% |
| Burgundy 150 | +1.9% | -9.1% | +3.1% |
| Fine Wine 100 | +3.3% | -6.8% | -7.4% |
| Bordeaux 500 | -0.4% | -12.3% | -18.3% |
| Fine Wine 50 | +1.2% | -10.0% | -22.4% |
Champagne 50 has the best five-year return of the eleven indices Liv-ex lists. Italy 100 and Burgundy 150 are the only other regional indices in positive territory over that span.
The two-year column is the honest part. Champagne ran hard into 2022 and gave a large share of it back. Trading Grapes describes the Champagne 50 in August 2025 as having "almost doubled between 2020-2022 before dropping sharply in 2023-2024", leaving it back near its January 2020 level. So the five-year win is real and it is also a round trip with a peak in the middle. Anybody quoting you a Champagne return without naming the start date is choosing the start date.
Liquidity held up better than price. Winecap notes Champagne at 12.4% of Liv-ex trade by value year to date in Q2 2025, against an 11.8% average across 2024. It also took the ten most recent vintages of five prestige brands, 50 wines in all, and found 43 had resisted price declines and 40 had been stable for at least six months. Winecap reports Pol Roger Sir Winston Churchill 2015 up 24.4% in the first half of 2025. A category people keep buying while prices fall is a different situation from one they have stopped buying.
Why peak-to-trough and five-year figures disagree by twenty points is worked through in the how to invest in wine guide. The live version of the same comparison sits on the market index.
Why the supply side carries the argument
Because Champagne sets its own volume on purpose. The Comité Champagne states that its executive board, made up of representatives of the houses and the growers, fixes a maximum yield each year inside an EU limit of 15,500 kg per hectare. It has cut that number in each of the last four years.
| Harvest | Marketable yield |
|---|---|
| 2022 | 12,000 kg/ha |
| 2023 | 11,400 kg/ha |
| 2024 | 10,000 kg/ha |
| 2025 | 9,000 kg/ha |
| 2026 | 8,800 kg/ha |
The 2026 decision, announced on 23 July 2026, allows around 250 million bottles. Set that against demand. The Comité Champagne reports 2025 shipments of 266 million bottles, nearly 152 million of them exported and around 114 million sold in France, slightly down on 2024. Harpers reports 107.1 million bottles shipped in the first half of 2026, up 1.2%.
The stock position is the number that matters most and the one nobody quotes. Vinetur reports that when the 2025 yield was set at 9,000 kg/ha, the region was holding 1.28 billion bottles, about 4.8 years of sales against a target ratio of 4.2. Cutting the harvest is how that gap closes. Maxime Toubart of the Syndicat Général des Vignerons put the logic plainly: "The goal is to adapt production as closely as possible to sales forecasts."
Two things follow for a buyer. Supply is being managed down by decision, which supports pricing across the appellation. And a small harvest does not mean a small release: according to Vinetur, 2025 production was forecast at 258 million bottles against sales of 270 million, with the difference of about 12 million bottles due to come out of reserve. Scarcity in Champagne is a policy decision before it is a weather event, and policy can reverse.
Which Champagnes actually trade?
A short list. Liv-ex builds the Champagne 50 from 16 champagnes, and fine wine trade overall is concentrated far beyond that: Trading Grapes reports that "in 2024, 80% of Liv-ex trade value came from just 2% of wines".
You can see the shape of it in what changes hands. Vin-X reports Dom Pérignon P2 2008 at £3,518 a twelve-bottle case in January 2026, among the ten most-traded wines on Liv-ex by value, alongside DRC Romanée-Conti 2009 at £175,492, Chave Hermitage 1995 at £69,534 and Screaming Eagle 2023 at £20,532. That is the company Champagne keeps at the top of the market, and it is a short list of labels.
Merchants tell the same story about their own books. Bordeaux Index says in its guide that Champagne "accounts for around 20% of its annual turnover", and points to "Salon 2002, Bollinger Grand Annee 2008 and Krug 2000 all doubling in price" on its trading platform. Read that as a merchant describing its own inventory rather than as an index, and it is still evidence that the category clears.
The practical consequence: Dom Pérignon, Cristal and Krug Grande Cuvée are liquid because thousands of cases exist and dozens of people want each one. A grower Champagne with 400 cases made and a cult following has scarcity and no exit. Both can be good wine. Only one of them is an asset.
How much does the vintage matter?
Less than in Bordeaux, and in a way that helps you. Champagne is a blended wine by default, and each house decides on its own whether a year is good enough to declare at all. So a vintage on the label is a producer's judgement, not a regional verdict.
Salon takes that furthest. Salon describes its wine as "the champagne of a single terroir: the Côte des Blancs; a single cru: Le Mesnil-sur-Oger; a single grape variety: Chardonnay; a single year: that of a great vintage". The house says only 37 vintages were produced across the whole of the twentieth century. Cristal is, in the words of Roederer's US importer Maisons Marques & Domaines, only produced in years when the grapes have reached perfect maturity. Maisons Marques & Domaines lists the 2016 as 58% Pinot Noir and 42% Chardonnay.
Then there is the release lag, which is the part most guides skip. The Comité Champagne specifies a legal floor of 15 months of cellar ageing for non-vintage and three years for vintage, both from bottling. Prestige cuvées ignore the floor. Salon says its wines are "aged in our cellars for an average of ten years, after which they finally start to reveal their complexity and finesse", so its wine reaches the market roughly a decade after the grapes were picked, into a market nobody could have forecast at harvest. That lag is why supply of a given vintage cannot respond to demand, and why the release price is set by the house rather than by the room.
Per-vintage detail for the region is on the Champagne vintage pages.
What goes wrong: buying at release
Release-price inflation. The houses raise the launch price of each successive vintage, and if you buy at release into a rising ladder you have bought the top of the house's pricing rather than the bottom of the market's.
Sara Danese documented this for Tom Hewson's Six Atmospheres in July 2023, tracking Cristal case by case. Cristal 2012 came out in 2019 at £1,670 a twelve-bottle case and peaked at £3,050 in late 2022. Cristal 2013 was released in early 2021 at £1,490 and rose 118% in its first eighteen months. Cristal 2015 was released at £2,729, an 83% increase on the 2013's release price, and had fallen 18% from release by the time she wrote. Dom Pérignon put its 2013 out 26% above the 2012.
Those are 2023 figures and the market has moved several times since. The pattern is the point, and the release schedule keeps producing it: the house prices the new vintage against the trading level of the old one, so a strong secondary market drags the next release price up with it, and the new vintage arrives with no margin left in it. The buyers who did well in that period were holding wine bought before the ladder moved.
What does a case actually cost to own?
Three lines beyond the price, and two of them recur every year you hold it.
Duty. HMRC lists £30.62 for each litre of pure alcohol on wine between 8.5% and 22% ABV, at the rates in force from 1 February 2026. It is charged on alcohol, not on value, so a £2,000 case and a £200 case of the same strength pay the same. Read the label and do the arithmetic: a twelve-bottle case at 12.5% ABV holds 1.125 litres of pure alcohol, so £34.45 of duty.
VAT. GOV.UK lists the UK standard rate at 20%, and it applies to most goods and services, the wine included once it comes out of bond.
Storage. Bonded storage is charged per case per year, and on a mid-priced case it compounds into a real share of your cost base over a five-year hold.
Buying in bond keeps the wine in duty suspension. HMRC states the duty point in Excise Notice 197, at removal to home use: When excise goods are removed from an excise warehouse to home use, excise duty that is due must be paid. Until then, nothing. The landed-cost calculator itemises each line against its own base, because duty runs on volume, buyer's premium on hammer, and VAT on several of the others. The full round trip, including what the exit takes, is in the wine investment risks guide.
Do you pay capital gains tax on Champagne?
In the UK, usually not, and HMRC's own manual is the reason. Two reliefs do the work, and the second one is widely misquoted.
The chattels rule is the simple one. HMRC's Capital Gains Manual at CG76901 says "Bottled wines and spirits are chattels (tangible moveable property) so disposals for £6,000 or less will be exempt". Sell to one person and the bottles may form a set, which turns on whether the bottles are 'similar and complementary' - which would require the wine in them to have been produced from the same vineyard in the same vintage year, and whether the bottles are of greater worth when sold collectively than when sold individually. A twelve-bottle case of one vintage sold whole gets one £6,000 limit, not twelve.
The wasting-asset rule is the one people get backwards. A wasting asset has "a predictable life not exceeding fifty years at the time when it was acquired". HMRC says that definition "would certainly not apply to port and other fortified wines which are generally recognised to have a very long storage life". For everything between cheap table wine and port, the manual's line runs the other way: "where the facts justify it, we would normally contend that wine is a wasting asset if it appears to be fine wine which not unusually is kept (or some samples of which are kept) for substantial periods sometimes well in excess of 50 years." Champagne is not fortified, so on HMRC's own wording a prestige cuvée sits on the exempt side of that sentence rather than the taxed side.
That is an internal manual, not statute, and it is HMRC's view of the facts rather than a ruling on yours. The statutory detail, and how the US, Germany and Ireland treat the same sale, is in the capital gains tax on wine guide.
How do you check a bottle before you bid?
Start with the disgorgement date, because Champagne is the one region where the producer often tells you. Vinfolio notes that Krug prints a six-digit code on the back label of every bottle. Vinfolio explains that the first three digits give the quarter and year of disgorgement. The system dates from 2011, so the code appears only on bottles released after that. Enter the code on Krug's site and you get the blend behind the bottle in front of you, down to the oldest and youngest wines in it.
That matters commercially, not just for interest. Two bottles of the same non-vintage cuvée disgorged years apart are different wines in the glass and different propositions in a portfolio, and only one of them has been sitting on a merchant's shelf since. A recent disgorgement on an old base is a wine with its cellar time behind it and its post-disgorgement life ahead.
Then the ordinary condition work: capsule intact, no seepage under the foil, labels consistent with the claimed vintage, and a storage history you can name. The full inspection sequence is in the condition and provenance guide.
Watch the bottles you want, not the whole region
You do not need a view on Champagne. You need to know the day one bottle you already want is offered below what it has been fetching.
Put the cuvées and vintages you are actually interested in on a watchlist and we will alert you when the price moves: a lot listed under recent realised levels, a release that has finished unwinding, a vintage whose drink window is opening while the market still prices it as young. That is the trade the release ladder keeps creating, and it only helps you if you are watching on the day rather than reading about it a quarter later.
Start on the Champagne producer atlas, which maps who owns what across the 319 crus the Comité Champagne counts in the appellation, then add the cuvées and vintages you would buy at the right number.