Champagne Is Holding 4.81 Years of Itself

Every wine region has a reserve. Champagne’s is different, because it is not a private inventory decision — it is an interprofessional instrument, agreed between growers and houses, and it exists to let the appellation decline to sell.

In 2025 that instrument was already full. The appellation is carrying 1.28 billion bottles. At EUR 21.42 a bottle that is EUR 27.42 billion of wine at current ex-cellar values, against EUR 5.7 billion of annual revenue: 4.81 years of shipments against a stated target of 4.2, or roughly 162 million bottles more than the appellation wants to be holding. Nothing else in the category’s balance sheet is this large, and no marketing decision can move it.

The yield did what it was asked to do

The arithmetic of a Champagne harvest is unusually clean, because the appellation publishes both ends of it. The yield was set at 9,000 kilograms per hectare for 2025, after 10,000 in 2024 — itself roughly a 12 per cent cut on 2023 — and the growers had asked for 10,500. Thirty-four thousand two hundred hectares at 9,000 kilograms is 307,800 tonnes of fruit, which on the Comité’s own conversion is 264 million bottles.

That is 1.17 kilograms of fruit per bottle. It is also fewer bottles than the 266.1 million that were shipped in the same year. The appellation released slightly more than one harvest’s worth and drew the difference from stock — and that difference is about 2.1 million bottles.

Hold that number next to the surplus. If the 2025 draw is roughly 2.1 million bottles against a surplus of 162.3 million, then at that rate the overhang closes in 77 years. The reserve is not a mechanism that clears a stock problem. It is a mechanism that holds one, and it was designed for a bad year, not a bad decade.

What the yield cut can and cannot do

The cut from 10,000 to 9,000 kilograms per hectare removed about a tenth of a harvest’s bottles — on the same conversion, roughly 26 million. Against a surplus of 162.3 million, one year of the appellation’s sharpest and most political instrument addresses about a sixth of the problem, and it can only be used once in a direction before growers’ incomes are the next thing in the room.

This is the structural asymmetry that makes Champagne different from a spirits category with an age statement. A whisky producer holding excess stock can simply hold it, and the stock improves. A Champagne producer holding excess stock is holding wine that has already done its minimum rest, in a bottle, against a market whose price is falling. The 2025 price series — revenue per bottle from EUR 23.95 to EUR 21.42, down 10.56 per cent — is what happens when 162 million bottles of good wine look for a buyer who is not in a hurry.

Why the stock exists at all: the ageing rule

Non-vintage must rest on the lees for at least fifteen months; vintage for at least three years. Every bottle sold therefore carries between one and three years of working capital before it earns anything, and the appellation’s own answer to that has always been to hold more wine than it sells — a buffer both of inventory and of blend options.

At EUR 27.42 billion of inventory on EUR 5.7 billion of revenue, the buffer has become the business. And the capital behaviour in 2025 says the industry knows it. LVMH cut operating investments from EUR 538 million to EUR 222 million — down 58.7 per cent — faster than revenue fell and faster than profit fell. Laurent-Perrier turned an operating cash outflow of EUR 11.2 million into an inflow of EUR 24.5 million, a EUR 35.7 million swing it attributes to the level of investment as much as to trading. Both ends of the ownership spectrum made the same decision: spend less, because the cheapest place to find a bottle in 2026 is your own cellar.

Three ways the surplus actually ends

One: it is discounted. This is the route with a price already visible. A blended fall of 10.56 per cent across 266 million bottles is not a change in country mix; it is a change in the terms on which wine clears. Every further year of overhang is another year of the same pressure, and it lands hardest on the participants with no brand to protect — the growers at EUR 17.52.

Two: the harvest fails. Thirty-four thousand two hundred hectares and a fixed yield set the ceiling; frost, hail and disease set the floor. A short vintage is the only event that removes bottles faster than the appellation can decide to. It is not a strategy, and a category that has to be rescued by weather has no supply policy at all.

Three: demand recovers. This is the one the forecast assumes and the one the appellation cannot control. Champagne’s volume is tied to the celebration occasion, the on-trade list and gifting. An appellation can ration supply into a stable demand; it cannot ration an occasion into existence.

The number to watch

Not shipments. Shipments can be flat for a decade in a category that has been moving roughly 266 to 271 million bottles a year. The number with information in it is inventory divided by shipments, currently 4.81 and targeted at 4.2.

If that ratio falls, Champagne is genuinely clearing and the price can recover. If shipments hold flat while the ratio sits above four and a half, the appellation is not selling what it makes — it is storing it, and storing it at a cost that shows up first as a lower price per bottle and later as a lower yield per hectare. The reserve was designed to smooth a year. What it is doing now is carrying a decade, and the interest on that is being paid in the one line item the appellation spent a century protecting.

Companion analysis: Global Champagne Market Report, Size & Forecast 2026–2033.